So you're staying you can still program as a hobby. Well, of course. You can also write poetry and do other things you won't get paid for. People are complaining that programming is being reduced to a hobby.
Yes. There is also reason to believe that the ecosystem is on the path to migrating to one where human developers are an afterthought at best. In the current phase we have lots of more stuff in markdown and other human readable formats and programming languages for sure but there’s little reason to believe that they will stay that way as humans get further from the code in general and the primary consumers and producers are AI. It would be strange if it remains optimized for English-speaking humans for very long.
Open source development has become buried in an avalanche of AI generated code as well so the character of a lot of hobby programming in that domain has changed drastically.
Given the amount of complaints and generally downtrending tone on HN over the past decade and a half, did people really enjoy programming for work? It feels like the vicissitudes of the market, from the “move fast and break stuff and technical debt be damned” era of ZIRP, to “move even faster because now companies have to show profit and we are post-ZIRP cracked 100x engineers” just led to a lot of sloppy inelegant code, features and products built that never shipped, and unending meetings about half-baked business requirements. And a lot of the programming just ended up writing API glue code.
If you’re a The Art of Computer Programming-reading, code golf-competing, blogging neat elegant hacks kind of programmer, were you really getting that experience coding at work?
Work always comes with bullshit. Its the nature of the thing. I don't think anyone was really getting code golf-style experiences at work, but they likely had enough positives that they outweighed the negatives, which resulted in net "enjoying" work. For a lot of people, AI removed a lot of the positives and left a lot of negatives, and some are still coping with that in one way or another.
I remember reading this New Yorker piece[0] when it came out 8 years ago about how the proliferation of electronic health records software in hospitals was driving older doctors towards retirement, and it made me ponder how many people hated using the software I've built professionally over the course of my career.
Change comes to every profession, and it's never universally helpful or productive. I'm very much in the middle of the road on the AI booster/skeptic continuum, but I know it's not going away.
And getting to "enjoy" the work was always just a perk. Where in labor law in any country throughout history guarantees enjoyment of the work? Surprisingly few people throughout history have been able to say they enjoyed their work. It's always been a trade of time for money, everything else is table stakes. I'm finding new ways to find enjoyment both in the job and out of it, and if that time-for-money trade stops being worth it, I'll figure out something else.
I always wondered why since the introduction of EHR systems do I have to fill out 5+ pages of personal information and medical history every time I visit a doctor, as if they have absolutely no records about me at all?
Interesting article. Another solution would be to try to reverse this trend, even the doctor in the article isn’t sure if things actually got better functionally by the end of it.
How would you reverse this trend, I dont know, but AI basically puts us in the same boat as the doctors in that article.
HN is 70% AI propagandists and bots who flood people's attention so they do not develop independent thoughts. This needs to happen daily, because 48 hours might leave people room to think.
These are talkers who have never produced anything, taken everything from open source while looking down on OSS authors. They one-shot a note taking app and feel they have done something culturally significant.
AI in software is the greatest farce since NFTs and attracts the same people. All web software got worse, nothing new is being written and everyone just talks about their precious conversations with the clankers like commenting on a football game.
I fear there is a tendency for some people who are against the proliferation of AI to make sweeping gut-reaction comments like this that do more harm to their position than good.
Perhaps everything that you said is 100% correct, but people are easily swayed by things that “sound right” being overly aggressive opens your position (in people’s monkey brains) to “enlightened centrists” that leave the door open for bolder opposition.
Many people have been genuinely helped or feel like they have been genuinely helped by LLMs in coding, they aren’t going to change their opinions based on fiery rhetoric and accusatory tones. Much like I can’t force myself to believe in Santa Claus regardless of how aggressively my nephews claim he is real.
I'm part of the skeptic camp, and I'd argue that we are never going to have LLMs solving all problems. Therefore, we need humans to know and understand how to solve problems. Unfortunately, when "the genie is out of the bottle" and most (but not all!) problems are solved, it becomes very very hard to argue for doing things the hard way.
Furthermore, it becomes practically impossible to argue for doing things the hard way for all of the solved problems, which were and are necessary for being able to learn the craft deeply. At the very least, deeply "enough" to advance the state of the art.
In short:
1) Humans are necessary.
2) Knowledge and experience are necessary.
3) Humans need time and practice to gain knowledge and experience.
When one side argues that one or all of the above points are false, I don't believe them.
Realistically, we are seeing people even in the most pro-AI groups begrudgingly concede #1 and #2, while expecting #3 to just sort itself out.
That’s the hard part with all of this, you’re either of the opinion that there should be some individuals capable of determining if the AI is doing the right thing, and therefore all the previous prerequisites are still necessary, or that we should just let it all ride on the AI, which is probably easier, but certainly more dangerous.
Tbh I think you are right. The problem is now striking the right balance, like we don’t need everybody to know how to build each part of smartphone or chip, but there should be significant group with that ability. This will be a huge challenge
really? i feel like the overwhelming sentiment in every comment section is staunchly anti-ai. it's far more common to see people lamenting their job prospects or complaining about having to review mountains of slop PRs.
I wonder what the ratio of poster sentiment to comment sentiment is. I figure the sentiment of a post is seen broadly as much more authoritative than the sentiment of comments.
for a long time, I was roasted on HN for seeing any capabilities in programing ai. Recently tables turned and more people see that ai is not leaving and will stay with us
I think it's definitely not accurate to say that these people have not made anything significant. It sounds like coping. How many programmers have made anything culturally significant? How many are just codemonkeys who fulfill someone else's requests, take their cheques, and go home?
In my opinion, the whole anti-LLM coding thing is purely ego preservation. And I get that, who the hell want's to believe their high paying job is going away? To say using LLMs cannot achieve results bigger than 'one-shotting a note taking app' is really just not true.
Comparing it to NFTs is absolutely wild also. NFTs literally were mass psychosis and had no value whatsoever. You're saying someone can't build software, even just for themselves, or otherwise, that doesn't have any value at all? How could that be true?
How many programmers have made anything culturally significant??
You are on an online forum geared towards programmers, using the world wide web created by a programmer. The concept of a video game did not exist before they were being programmed.
You think the world just appeared as it is and everyone only does things for money? Never because, unlike people who use every shortcut possible, they actually care about the details?
Basically, LLMs have reduced the incentives in doing open source.
If you go with agentic route, randos with money to spare on latest models, beat you. If you go LLM-free, AI corpos say "all the better for us! More free untainted dataset!"
I'm a perfect example. I'm nearing the first functional release of my entirely handwritten Linux-native mod manager, and I'm not sure if I want to open source it or not. I love the concept of FOSS software, and it's nearly the only kind I use when possible, and I would love to give back to other humans. However, I'm not interesting in doing free work for megacorps that simply want to see me out of a job for their benefit, and open source is currently looking more and more like that.
Humans are so lazy and obsessed with cost-cutting, efficiency, and "line go up" that we'll willingly walk ourselves into such a bleak inhuman future.
I fully agree with you, and especially on the 'temporarily' part. The good thing is, they can one-shot as much as they want, but almost any fully LLM-generated project that pops up still has inferior code quality compared to those projects from people who actually know what they are doing, so people who prefer code quality will ignore most of the AI spam anyway, just like we do. People deploying one-shot projects out there also have a very notable lack of common sense, skill, and knowledge on topics like security and/or code optimization (obviously, because they don't code), which is mostly obvious both in their way of argumentation and their published artifacts.
They are happy to have their ads-maximized agreed-to-all Cloudflare/Anthropic/OpenAI/Google/Apple/Microsoft accounts so they can one-click deploy that "worker" with all the UI gears spinning automatically, because who needs to know how to configure a web server anyway? :^) And wow, have you heard that Cloudflare offers an operating system now??
I don't know that anyone suggested it wasn't OK to enjoy programming.
My problem with AI is that it now excels at the part of the job that was my favorite (writing code) and it leaves me to do the part that I suck at (being a manager). In 18 months or so my job satisfaction has gone from pretty high to abysmal.
Writing code was the "downtime" for me, it existed between boughts of intense problem solving. With AI agents my time is shifting heavily towards solving problems and I have less downtime as a result.
With LLM and current economic situation my job is doing 5 things at once and after 7:30h my brain is just in panic mode. On remote this isn't a huge issue but in the office (which I'm obligated to be twice a week) I just don't know what to do so I eat 3 meal 5min each pretending this is normal.
Then the next day after this brain haemorrhage inducing experience I just waste 2h remotely on gathering mental capacity to simply know what next I need to do and next 2 says I repeat until weekend.
Simply because my boss is developer in love with LLM and machine gunning 20 PR daily.
Not to mentioned PR with more code than I can reasonable can understand and say anything about it in a project I didn't touch.
So, currently I can only say I hate my life. I'm looking for an idea which pays similar but is as mindless as humanly possible
"Gathering mental capacity to simply know what next I need to do" is not a waste. That is work. In fact it is the most important work with LLMs. If you spend all day in a hammock dreaming up what next needs to be done (and it's the right stuff), and then type a few sentences into the LLM to generate it overnight, that is better time spent working than spending all day cranking out 20 PRs that are of mid-low value.
Yeah, if you're wandering in random directions all day long (20 PRs a day is rarely working towards a clear goal) then you're not going to get far. Taking some time to figure out where you're headed is probably way more beneficial in the long term. And it's also something LLMs suck at, they interpolate extremely well but haven't got a clue what's next.
Do they know they are, essentially, expending you?
It seems like your boss is doing work that doesn’t actually add value without you to review the code. But maybe the company isn’t set up to fairly evaluate the fact that review provides more value than creating code, with a workflow like your boss’s?
> Do they know they are, essentially, expending you?
I don't about the OP's specific situation, but there are definitely managers and management consultants that consciously try "expend" their employees.
Jeff Bezos is a famous example. IIRC, to avoid "complacency," he wants constant turnover of employees. Hire new people, burn them out while they're new and enthusiastic, then replace them. That's why it's occasionally leaked that Amazon consider such-and-such employment market mined out (e.g. everyone's either worked for them or is wise to their ways), so they have recruit further afield.
Maybe this will provide some historical context. Capitalist automation is an anti-worker, anti-union ideology even though in theory automation is a good thing. When automation is used to reclaim time for workers or to increase pay, it is a good thing. When it is used to dispossess people or enforce work speedups, it is anti-worker.
I don't know your full situation, but as someone who has gone through full-blown real burnout: it's not worth it. In this economic situation, we do need to grind to keep our jobs but if you're getting close to burn-out I would *strongly* advise prioritizing avoiding burnout.
I went from being a passionate programmer spending all my free time working on code, to be unable to write a single line for even my most loved personal projects for 6 months. Fully awake, in the middle of the day, I would boot up my computer and within 10 minutes my body would put itself to sleep. I'd wake up hours later with my laptop barely past the login screen. It took me 6 months before I could do personal projects again. 1 year before I could do professional work again. 2 years before I could juggle both. I've fully recovered (or at least I did right before LLMs hit the scene), but it takes a lot of time and I lost a lot of money not working for that year. If I could go back in time, I would have quit that job earlier because it is far easier finding jobs as a passionate developer rather than a recovering burnout.
I think the fun part of stuff you could do is setting up dev environments that are augmented. Not only the code creation part. But also the reviewing, testing , quality benchmarking and more. Arranging this is not easy and definitely an engineering challenge.
I am surprised how many people like the actual writing part, it always felt to me like doing micro in an RTS game. I know a lot of people love that too but I always played RTS games to feel like a general. But I'm still in the camp of reading every single change the LLM does, so I feel like I get the best of both worlds, but maybe that will end up being a waste of time soon. Like if the job ends up being just writing and reviewing MD files for the agent to digest I'll hate it too.
It's funny how differently people see things. I find reading/reviewing LLM generated code to be much more difficult and tedious that writing it myself.
Sure, I guess, but I've yet to see an LLM generate code that I'd consider either easy to read or particularly well-written, regardless of who was driving the agent.
And even if it does end up being good code, if it's something non-trivial, having a reasonably deep understanding of what it's actually doing is quite hard simply from reviewing the code.
I just don't get how you can properly reason about the code if you're not involved in writing it, though.
I keep giving AI-assisted programming a serious try, but time and time I land in the exact same place: reading the code is more exhausting than writing the code, _and_ I end up with a far worse understanding of what is being built than if I had just written it myself. The fact that LLMs tend to write _way_ too much code certainly doesn't help.
And what makes me sad is that, if you follow this line of inquiry, you eventually realize that capital isn't challenging the value of humans _writing_ the code, but rather the value of humans _understanding_ the code. Because, as I've experienced many times when talking to those how have completely bought in on the hype, the answer is always to just "let the AI handle that." AI writing too much code? Have it summarize it for you. Your understanding of the code is worse than if you wrote it yourself? If it works, who cares!
And yes, I think we are well on our way to becoming Markdown jockeys.
Ever since AI reached a certain level of quality, I haven't been able to stop (vibe) coding. I've discovered that I really enjoy designing and building products, more than tinkering; occasionally, I come across algorithms or other technical details that I find interesting, and I switch to manual mode.
At some point in my career, I just got tired of having to read tons of documentation for fragmented libraries and frameworks, and AI is a "solution" to that.
Cynically, this could be seen as plumbing, but there's an art to designing efficient and usable products, that I personally enjoy very much.
I phrased it fine. No-one knows your intentions. If you don't think it comes off as insulting then feel free to disregard, but I (and others, apparently) think it does.
I've seen plenty of people here claiming things like "anyone who enjoys coding hates AI," "anyone who enjoys the process as much as the outcome hates AI," "anyone who's creative hates AI," etc.
It's all projection to make themselves feel better about their difficulty adapting to change.
I've always enjoyed hacking on random projects and AI does that really well. Every crazy idea I've ever had can be iterated on, and takes way less time now. Sure it's awesome that I don't have to spend hours throwing together a bunch of python and bash scripts, but it just happens instantly now. That's why I actually like Copilot autocompletion for side projects, because I can build something and still save a massive amount of time tinkering with limited blockers. I wish there was a better middle ground between "one shot this" and writing everything yourself.
> My problem with AI is that it now excels at the part of the job that was my favorite (writing code)
My problem is it doesn't even do that. LLMs still aren't good at writing code (despite the hype) and you have to review everything they produce, which takes just as long as writing it myself would. But the industry is falling all over itself to use them to write code, doing a worse job than humans would. It's extremely frustrating.
I agree it’s very mixed, it strongly depends on what you are developing, how big and established your codebase is, how opinionated you are about simple vs overdone, how many subtle and nonlocal interactions your codebase has, etc. Plus the more experienced you are the more you are likely to reject the AI solution as not well considered, whereas to someone new may think it looks plausible and confident.
They're great and you don't have to review everything.
From what you're always commenting, I could believe that you were exclusively using Haiku 4.5, which would explain how the tool produces only garbage while the rest of the world pretends AI was great.
Just out of curiosity, do you use like Opus 5.5 in Claude Code, and you still find the results inadequate?
they're about as great as an average developer. See rsync fixing their fabled code for some time now. They also have more output than the average programmer. so...
I’m not the commenter you’re replying to, but IME with the recent models (like Opus 5.5, Fable, etc) - yes, still inadequate. As a sibling comment states, it’s average code. Which also makes complete sense.
That bar is too low for me
I am on the skeptical side of LLM performance (why is 2026 “the” year? Why not in 6-18 months? Is it clear that they’re really hitting all the benchmarks for coding ability?) but was industry code in say 2022 really all that much better?
Yes, they're very good at writing code now. The problem is that we (or at least I am) drowning in code that runs and does that resembles the problem you asked it to solve, but that something is frequently semantically incorrect.
Something has to give; the constant onslaught of slop and workflows changing every few months just isn't sustainable. Either we slow the fuck down and figure out what we're even doing here, or the AI improves so much that we're all out of a job.
Look, people are going to pile on and talk about how much they enjoy coding now.
Those people have processed this change differently or sooner or later than you have.
This is ok. You will process the change, and you'll come out stronger for having done so. It takes time to decide what you want to do in the world, and even more time to decide how to react to the world changing and moving away from your planned path. And it takes even more time to actually see if the world is moving in a way incompatible with your plans.
It will be tempting to cycle into negative predictions and negative self-talk. This will not help, and will not be productive.
You, ryandvm, built a career around something you love, and that is amazing. You can choose to do it again, to find something you love, to be stalwart about how you work and who you work with, or any number of other options. The thing that nobody can change is your agency.
I struggle with the nature of work every day, and have for months. It has caused me to re-evaluate what I work on, why I do it, how I structure my finances, where my kids go to school, literally everything. Still the ship turns, as it must.
You sound like you're processing this well. I think I'm in the group of people you're talking about who aren't processing it well. What kind of conclusions have you come to regarding that list (what you work on and how, finances, kids, etc.)?
* I made the classic mistake of assuming the present was permanent. It never is anyway, small changes always ruined the present. The accumulation of small changes over my career has changed the way I work probably as much/more than AI has changed it this year, if you can believe that.
* I discussed the family budget with my wife, and made a plan to halve our income requirements. It will require a race in the short term, but allows a coast later on. This is key for me - almost all my tension regarding this phase is about the future earning potential of our household. If that goes away, who cares how work changes?
* I made the other classic mistake of not remembering how much I have changed and adapted over time. The breadth of things I have worked on is pretty remarkable, and the things I have learned to do are crazy. While things are changing fast now, they aren't changing quite as negatively or immediately as some claimed they would. There's room to adapt.
* I stopped assuming I had to work the way people said I should. Aside from CEO mandates, there's a ton of wiggle room in how you produce code. I read changesets and make manual adjustments to the pattern, etc. I write my own parts when I want to. I specify a design that I want rather than asking for whatever the AI wants. I get to own everything but the typing that way, and my RSI is improving dramatically b/c of it.
* I use AI to set up programming challenges when I feel disconnected. It sets up a tangible problem, and acts like a coach only when I need it. I use this to preserve the feeling of sharpness. Earlier this year I challenged myself to write, in C, from man pages only, a "ping" and "traceroute" clone. I did so on a flight back from a work trip.
* And I started reaching out to friends to hang out more. It's hard to doom about AI when your friends and you are having a beer and playing Container or Captain Sonar. Life has always been what's "out there" not what radiates from your monitor.
From a recent post here in HN, my agent stored it without the URL it exists just as text for me to get philosophical with, I do that with insights such as above, so here is a quote:
Derek Thompson, in "I Know Why the Mathematicians Weep": "If intelligence becomes less uniquely human, we may come to value more highly what remains uniquely human: our capacity to relate and to care for one another".
Funnily my bot was called Cladius van der Bot, nice name ;)
i dont meant to come across as too harsh, but that is something that you will have to figure out. if companies are less inclined to give salaries for technical skills, then its on us to figure out how our current skill sets become marketable.
even the job of freelance engineer is customer relation. how good are you at extracting product requirements? understanding a clients workflows? these take relational skills and are often what freelancers are paid for, beyond the ability to produce code.
There are actually a lot of professions that either directly or indirectly rely on that. For example, therapists and social workers of many stripes, actors, writers, musicians, etc benefit directly from that, for at least some sub-disciplines/genres. Indirectly, almost every job where you interact with others directly to provide a service or sell a product (which is many, many things, and more than might first come to mind).
I doubt most people will be developing a rapport with an AI agent trying to sell you that contract at work, or helping you when you're irate and have escalated a problem a few times.
> There are actually a lot of professions that either directly or indirectly rely on that. for example, therapists and social workers of many stripes, actors, writers, musicians, etc benefit directly from that, for at least some sub-disciplines/genres.
All of those except for therapy are awful career fields. Musicians, actors, artists (outside the tip top of the field) are all the butt of jokes about how broke they are. Social work chases people out once they realize how powerless they are for such poor pay.
Therapist might make out well for a bit, but it seems like a loaded term these days .
> All of those except for therapy are awful career fields. Musicians, actors, artists (outside the tip top of the field) are all the butt of jokes about how broke they are.
If you're aiming for the top, sure. There are plenty of people who make a living doing stuff like that and are happy because they enjoy it. You're unlikely to be one of those people if you move to an area where there's an extreme over-supply and try to "make it big", or try to be "pure" to the art and eschew commercial options. e.g. An actor in LA, the starving artist in a loft in NY.
> Therapist might make out well for a bit, but it seems like a loaded term these days .
Loaded how? I was thinking of psychologists, licensed mental health professionals, physical, occupational, and speech therapists.
All of what you said is wonderful, and I want to particularly call out this:
I stopped assuming I had to work the way people
said I should. [...] I get to own everything but
the typing that way
I think this gets to the heart of my disconnect with the "I miss coding" crowd. I still get to do most of the stuff I found fun w.r.t. creating software.
Typing the actual lines of code was fun, sometimes, but it was never the goal. I feel like I'm still solving the same problems.
I agree 100%. I did enjoy writing code, but at the end of the day, it's the building/designing/problem solving that I enjoy. Whether it's related to software, woodworking, small engines, etc.
A non-zero amount of people are preparing (or have already) to leave the industry entirely.
Even if one were to continue doing everything by hand, it will still be harder compared to others who aren't. As in, companies will come to expect more/faster work than you are capable of, so it still affects us old farts that don't want to change.
We'll find something else, sure, or just retire... but not everyone is/was ready to.
I'm not sad on a personal level per se, I can still be positive and look to the future, but I lament what we increasingly no longer have.
I too have lost something I loved, I gained it after years of relentlessly focusing on maximizing what I loved and minimizing what cost me energy. And now the world flips on me.
The good news? I got pretty good at consciously finding the energy givers and drainers and optimizing for the givers, so I'll just keep doing that. I'll either start enjoying micro-managing LLMs, or find something else entirely. Or find that the LLMs need less and less micromanaging, or start enjoying whatever they need next year. (Point an army of agents in a direction?)
I came from the molecular biolab with Linux as a hobby, then when to software dev and infra. What's next ? Idk, I still love biology, keeping eyes and ears open...
> You, ryandvm, built a career around something you love, and that is amazing. You can choose to do it again, to find something you love, to be stalwart about how you work and who you work with, or any number of other options. The thing that nobody can change is your agency.
I've always taken the view that it was luck - lucky that my interests aligned with the market, lucky the market put value in what I am interested in doing, and lucky that I was given the chance.
I was never in it for the money, I was always in it for the enjoyment, and I will remain in it for the same thing.
The market for my knowledge and skills might come and go, there might have been a brief moment in history when someone like me could make a living off these skills, but like so many other creative careers that time was always going to be fleeting
Really wonderful comment. Thanks for sharing. I've noticed myself spiraling into doom-and-gloom scenarios about the future of this job (and therefore my future), and I've found it really unhelpful and really unproductive, in the sense that it sort of narrows my mind into a tunnel vision of decreasing possibilities.
The job description fundamentally changed to the point where we're expected to do something very different.
No amount of processing will move someone from ordering a steak to just enjoying their given peanut butter sandwich instead. They ordered the steak, and now the restaurant is telling us all too bad there's no steak anymore - eat your damn sandwich.
I don't think the metaphor works because a job is an agreement to exchange labor for money, not a service or a product that you purchase. But for the sake of argument, let's grant it. If a restaurant gives you a peanut butter sandwich instead of a steak, you can go to another restaurant that provides better service. Similarly, if an employee doesn't like terms set by their employer, they can move to a different company, or start their own company on their own terms.
To quote the lovable DHH, "writing code by hand is no longer an economically viable activity"
To adjust the metaphor a bit.. the cows have revolted and we can no longer get any steak.
If writing code via AI is an order of magnitude faster there won't be companies hiring people who want to code by hand. If you try to start a company by doing so then you'll lose to competitors without that restriction.
The choices left are: be homeless, change careers completely or suck it up and do something you don't like after being career-catfished
When you think through the 'change careers' path it also sucks and might not even be viable for many
I think this is basically the right attitude. The focus should be on what you really value intrinsically, and what changes you want to see in the world. That determines the goals that you set, which determine how you choose what to learn and what to work on. There's no set "career path" you should follow to get from point A to point B.
Am I insane? I could never imagine speaking to somebody like this in real life. I'm not saying this was ill-intentioned but there's something about this comment that comes off as REALLY weird to me. It's like religious/cult-speak.
You're not insane, because I used to get a similar feeling when I was first exposed to this kind of discourse. It creates cognitive dissonance in people like you and me, because you can clearly see that it was well-intentioned and grounded in the idea of empathy, but it ends up lacking empathy in ways that people like us consider important.
The comment you're referring to aims to be helpful and empathetic, but the fundamental problem with it is that it essentially dismisses the very real problems people are raising. It tells you that the change is inevitable and that you shouldn't and/or can't fight it. It tells you that what you're dealing with are your own feelings and, although they're "valid" (in the sense that it's understandable and human to have them), you need to work on overcoming them or finding a way around them. And if you can't, then you should either move out of the way of the progress or you'll remain unhappy.
It's an incredibly frustrating way of being talked to, one that I've seen managers employ quite a lot. I call it the "management aikido", because it takes your own impetus and turns it back on you, by reframing whatever issue you brought up to put the onus on you. Most of the managers I've seen use this technique believe they're doing a good thing by "empowering" you to deal with problems. What it does, in fact, is discourage you from raising issues in the first place.
On top of having the same "management aikido" vibe, @jvanderbot's comment ends up being "socially insensitive", for lack of a better way of describing it.
Over the course of history, we've seen great shifts -- typically tied to technological evolution -- that caused significant suffering to a group of people but (arguably) ended up being a net positive for our society. And when we look back, we inevitably feel a sense of empathy for those people and we recognize that what happened to them sucked and we express the hope that maybe next time we could do better as a society.
And then it happens again, like it's happening now, and we not only fail to do better as a society, but we talk like trying to do better is unreasonable.
That, in essence, is what makes @jvanderbolt's comment sound "religious" to you: the unquestioned assumption that what is happening is supposed to be happening and that it's you who has a problem that you have to work on.
Yeah I think you nailed it. I also think there's a degree to which we pretend victims of economic injustice aren't "real", for lack of a better word.. If the original comment was re-framed around vehicular manslaughter or rape, I think people could more clearly see that it isn't really empathetic. That's why the sheer reaction that it got from so many people kind of blew me away.
I'm definitely stealing "management aikido" by the way, thanks for that. :)
You could divide "your future" into two paths, conservatism (not the US amalgamation) and "missing the times that'd been", or continue move forward and adjust as the world change, which it always will, both in positive and negative directions. Things are changing, this is hard to disagree with, even if people disagree how things are changing, most people probably need to rethink a lot of their assumed perspectives, as I too had to do, faced with what I believe to be new realities.
I don't feel like it's patronizing to remind people about achievements they've accomplished and that they're still a capable person. Nothing we say to each other here on HN should be taken as "This is how you must live you life" but more like "Here's is something I hope helps you, if not, no worries", as we hopefully assume good faith out of each other.
That wasn't my intent. The intent is that the prevailing pushback that coding is more fun now doesn't apply and need not. That OP has to come to terms with that they want from the new world, even though nobody can say for sure what they world is, other than "currently different"
The thing about this debate is that "writing code" is not at all an atomic task. You need to actually type out code of course, but also edit, think, analyze, design... (not meant as clear or non-overlapping categories; but you get the idea) and you might feel very differently about some of those things than others.
> My problem with AI is that it now excels at the part of the job that was my favorite (writing code) and it leaves me to do the part that I suck at (being a manager).
I love solving problems, programming was always about that for me. The sucky part was always "I know the solution and written it out, now I need to use a text editor to actually make it work like that", and the fun part was always figuring out how to solve the problem, making sure it actually does solve it in real-life, and once it's been implemented, iterating on it until it gives people some sort of pleasure to use.
Most of the other passionate programmers I've met in life have thought about programming the same way, all of us favored approaches with less code, not more, because typing sucks and is slow, and more code means more bugs. I haven't met any of you "I love typing special characters into a computer" people in real-life, and I'm wondering where I could, because clearly it isn't just you who feel like this, seems there is a somewhat large section of programmers who enjoyed the typing experience more than the problem solving part of programming.
What's always been clear, is that people get into programming for different reasons, and some of those reasons might no longer lead to the lucky life of working with what you love and get paid handsomely for it, which sounds terribly sucky and scary, and I do feel for the people that now have to face this.
I can't speak for anyone else, but I've never really clicked with a "plan everything upfront, and then all that's left is writing the code" process
I've also never really experienced that professionally. I started as a professional programmer in 2012 and I've always been pushed to do some form of "agile" where you figure stuff out while you work because building detailed specs upfront is too costly and time consuming
So now I'm being told it's fine to pivot to these AI tools and it's awesome, you just plan upfront and execute. But I genuinely don't think I have those skills. My workflow has always been "have a rough idea where we're going and fill in the blanks as we go"
So now I'm trying to adjust to this new work, planning upfront with more detail and I'm struggling.
Coding is how I used to work through problems. I was really good at it. Now I feel like a fraud. Just a hacker, not a software engineer. I don't know how to get to where I need to be, and I'm really unsure I even have it in me to learn and entirely new way of being
I went from feeling on top of my life, my career on a great trajectory, and really confident I could build anything that anyone needed, to feeling completely worthless
And there's such little empathy out there for anyone going through it like me. AI is here now loser, keep up, get on board or get replaced.
If you read this far thanks. And if anyone has any advice for me, how to go from "thinking through problems by coding your way through them" to "doing the thinking upfront" then really, I'm all ears.
> And there's such little empathy out there for anyone going through it like me. AI is here now loser, keep up, get on board or get replaced.
I think that this really needs to be internalized by a lot of people. I really appreciated when my boss sat our team down and said "Yeah this is the new normal, and I know its not what you signed up for. Its fine to feel sad that your job is changing in such a dramatic way." No, it didn't help anyone actually adjust, but sometimes you just need someone to empathize and tell you its OK to feel bad about this when everyone around you is excited.
> If you read this far thanks. And if anyone has any advice for me, how to go from "thinking through problems by coding your way through them" to "doing the thinking upfront" then really, I'm all ears.
I'm trying to replace it with "thinking through with AI" to limited success. A lot of people recommend having the AI "plan" the thing and review the plan, which is OK. I've also had it do a prototype and poked around a bit, summarized the learnings, and then did it again in an iterative manner without rough edges I've found. Whether you are given the space to do that is up to your org -- its hard sell to "do it again" but I've found that the second attempt leads to better results since I can prompt my way around the problems. Sometimes this also leads to a "Not even wrong" moment where I found out that I framed the problem wrong and the AI dutifully carried out doing the wrong thing.
I've commented elsewhere about this but so many are flippant towards your feelings and it's quite sad.
Comments such as "now that economically 'traditional coding' no longer works" get over it. Or a recent one "get over the anthropocentric view of intelligence AI is here now".
These views (and many more besides) are short sighted, I do think anthropocentrically. The economy is a proxy for the lives of real people, the economy matters only because it deals with real people. People who feel, have agency, who live and die.
Two super AI's talking to each other maybe the most intelligent things in the universe but they don't matter.
Humans matter, human lives matter, human feelings matter.
I don't have a good answer, but a question of my own. I wonder if there are only the two extremes: write everything manually or one-shotting without looking at the code?
One-shotting without review is very quick but has its downsides: burnout, less understanding of the code base, less interacton with your colleagues, you miss the poor choices the LLM made so the code base might rot.
Writing everything manually is much slower and in some circumstances not what you want anyway.
Is there a middleground where you get significant parts of the productivity/speed increase but retain the most important parts of the manual process? I have found I preferred the copilot style of working, but the interface was not really optimized.
Instead of spewing out code at 20x speed can we settle for 5x and remain in control? That would be my ideal and more sustainable, right?
I wrote a thing about this (https://www.klio.org/theory-of-llm-dev-skepticism/) back at the beginning of the year, in a totally different era of AI dev tooling. What you're describing is what I called there "bottom-up" programming.
Unfortunately, I don't have a lot of first-person advice for you; as I say in that essay, I was also a bottom-up programmer for a long time and was deeply skeptical of top-down approaches. What changed my opinion and my practice was just doing things for long enough that I started having a sense of where things were going to go before I did the exploratory work to find it out, and eventually the exploratory work was unnecessary and I could create a basically-correct top-down design early on. I don't know how you build that experience without actually getting in the reps (which has some implications for future generations of developers, if we have any).
That said, my second-person advice would be to use the agents for the design part of things, too. "Here's the problem, how would you design it?" and then have a conversation with them, push back on the things that seem wrong, imagine reviewing this as a finished design from a coworker and seeing where it looks right and where it doesn't. Ask why it designed things a certain way; suggest alternatives for it to evaluate and think about what it's weighing. Basically borrow the robot's experience to augment and build up yours.
Actually, most of the stuff discussed in this thread are all ideas that those old-school software scientists already talked about long ago. Go look up the papers Bélády wrote back in the day.
Well, I can't speak for the person above, but in my case typing out the code is part of solving the problem. Plenty of times I've thought I had a great idea in my head, and then in the process of writing it out found better understanding and an even better approach. Some of that can still be emulated with LLMs -- you can read the output and have it tweak things where you find oddities, but its still not really the same, and I miss out having that process of discovery and improvement.
It really depends, it seems. Through my coding career I always had interesting, and maybe a bit too big ideas for consumer software, bogged down by the fact that I like to code interesting algorithms but not business processes and interfaces. And in that sense AI is really a miracle for me - I no longer need to get bogged down by tedious work, I can focus on architecting the software and interesting algos (nothing seem too big nowadays) and have AI code it, debug it, and deploy it in a fraction of time it would take me if I had to code it entirely! Also, I no longer need to learn all these modern frameworks that keep spurring like mushrooms. In fact, frameworks are not a necessity when AI can vanilla JS everything anyway.
It is an amazing time to be alive, scary and uncertain, but amazing nonetheless :)
> I don't know that anyone suggested it wasn't OK to enjoy programming.
Outside of LinkedIn Lunatics and social media engagement bait, I agree that nobody is saying you can't enjoy programming.
Of course we can all still enjoy programming. I still do some personal projects manually. It's fun and it also keeps my skills sharp.
It also makes you realize how much non-optimal stuff slips into entirely AI written codebases if you're not paying attention to the code. I can push on Opus 5.5 or Astra all day and get output that works, but when I get into the code and start processing it myself I ask a lot of questions that result in "You're absolutely right!" responses followed by having the LLM clean up what it did.
The sad part is knowing that the golden window or programming job enjoyment has closed. We had a relatively small window where software jobs paid extremely well to do work we enjoyed. My happiest career memories were working with small teams at startups where we would sit in a room together and be focused on coding for hours.
I'm sad for a second reason: The years before AI took over, my jobs had been trending away from that focused coding time for other reasons. The trend in the industry was to stuff companies full of roles like product owner, scrum master, cross-functional project managers, and other roles. These are highly valuable in small quantities with the right people in place, but at some point it got out of control. Most of the companies I was with felt like a constant circus of meetings and debates over what to build, with little time left for that focused coding work anyway. I still cringe when I remember a meeting with a CTO who explained his target ratios for engineers to designers to product managers to engineering managers to project managers where we realized software engineers were becoming a small minority on the software engineering teams. I disliked it at the time, but now I'm extra sad that my last window of possible focused paid-to-code time was spent mostly in meetings and Slack debates.
Part of the problem is that senior management in many organizations is so damn bad at, well, organizing the organization. Both the people and the software architecture. There really is something to an approach like everyone credits to Amazon: small team with a discrete product and proper architectural boundaries, communicating to other applications in known predictable ways with a proper process on both ends for making changes.
When you do that, a team can just go. Instead, in many places there's all this bullshit about "Application X is built by Team A, but Team B and C will do the QA." Or "Team D is in charge of Applications Y, Z, and Q, so they can't get to your feature request for Q this PI, because they've got high-priority work for Y and Z." So now you've self-blocked all work on Q just by screwing up your org chart. And we haven't talked about the fact that Team A is in the US, Team B is in Europe, and Teams C and D are in India, so now let's add time zone nonsense to all of the above stupidity.
You don't need as many of those ancillary roles if people involved actually understand Lean and how to run an SDLC, but in my experience, there are too many devs who don't care about anything other than slinging code and too many managers more interested in expanding their own rice bowls and looking important than in actually streamlining how stuff makes it to prod. And then you need a pile of PMs or Scrum Masters to sort out the mess.
It also leaves the part of reviewing the code (which I have realised is really quite different skillset from writing code - it needs a sense of something is fishy here or why is this even here or can't this look better).
It may be obvious to conclude that good programmers are good reviewers but I think all variations exist. I wonder how many people don't like their jobs anymore because they don't like reviewing code written by AI.
I’m in a similar boat, though I’ve found some solace in that AI development (for me at least) has also made other kinds of work more enjoyable. Previously, since I enjoyed programming, problem solving, and learning more than the end product, I tried to work on projects where I would learn as much as possible which had interesting problems to solve that would make me a better software engineer.
Now I can work on much larger and much more challenging projects where I care more about the end result than learning little details.
Something has certainly been lost, but I’ve at least gained something else in return. I can also do use an empirical approach now, like testing whether a big change would be beneficial without having to map it out and try to estimate the effect. This has made what I try to work on more research-like than engineering-like; I’m trying to figure out if something is possible rather than how to best implement something I know is possible.
I still worry that soon there won’t be any way to contribute economically, that the machine will eventually be better at everything, but this is a much broader problem than just software or even engineering.
If anything, I enjoy programming more now with AI assistance, not vibe-coding. The depth and breadth of my work has increased greatly (yes, I still read and write a lot of code and documents, that the agents build on); I'm like a kid in a candy store.
Projects that would have seemed too daunting now seem feasible to tackle. I could even get started on implementing my dream language, of a knowledge management system for agents, or ... The possibilities are endless.
Now, my software engineering is for my own company, maybe that makes a difference. There is no one breathing down my neck to use more AI where it does not make sense, or to use AI this way or that. I am in the driver's seat, and I only delegate to AI what I need it for, and what it is good at. There is still much that is required of me that the AI cannot do, not anytime soon.
> Now, my software engineering is for my own company, maybe that makes a difference.
That is 100% the difference. Most people working as employees aren't passionate about the company as an entity. They won't love your baby like you do. What they did love was taking your baby for walks; now they're left with just changing dirty diapers.
Every day I have this glorious schadenfreude on the software industry. The very industry that's been relentlessly rent-seeking, while eliminating jobs, after telling everyone to "just learn to code" is going to zero. In the past month I've replaced nearly all monthly software subscriptions with bespoke clones that meet my needs better. The only stuff I pay for is because there is data that I cannot acquire myself (i.e. live and regulated market data). The snake will eat its tail.
For what it's worth, I also love coding and have been coding since I was 5, but the writing is on the wall. The stuff I get done in a day now would have taken man-years before. It's like having a genie at your disposal.
> My problem with AI is that it now excels at the part of the job that was my favorite (writing code) and it leaves me to do the part that I suck at (being a manager).
I have always been the opposite, despite mostly having IC roles in my career. Still, I don't enjoy the change, mostly because the UX of this tooling is absolutely miserable when you need to do anything serious with it. I am past the point where I thought it was user error and think it's just fundamentally baked into the ecosystem to be annoying to work with. At least in most management situations, the employee isn't going to gaslight you, question every tiny decision you make (or worse, subvert ones it finds more 'favorable' for itself) or just make up wild, easily disprovable bullshit in a constant stream of verbal fecal matter. I've worked with maybe one person that came remotely close to that and they were easily ignored. This stuff's persistent and unavoidable.
I am getting a lot more done, but I am a lot more tired.
Conversely for me AI is letting me focus on the part I actually enjoyed: organizing and planning systems, and powers through the parts which were just a grind to get there.
By far my biggest use at the moment is just getting all the software I use to function as I want it too.
I think blog author's point was that even if AI is faster and better at coding, it's still okay to write code yourself - at least as a hobby - because that's something you enjoy.
I'm much more on your side personally though, because first, when you know there's a better and faster way to do something it does take away from the enjoyment of doing it the old way. Some of the enjoyment for me came from being unusually good at something vs. everyone else.
And second, even my hobby projects are always something that I want to make real, more than they are something that I want to do just for the fun of working on them. Even though I enjoyed the programming and puzzle solving process, the end product is still the main goal. So I'll be using AI at home as well (and it is nice to reify projects faster than before).
Having said all that, I recently read Use Of Weapons (1990) by Iain M. Banks and this passage encouraged me about the future:
"Can’t machines build these faster?" he asked the woman, looking
around the starship shell.
"Why, of course!" she laughed.
"Then why do you do it?"
"It’s fun. You see one of these big mothers sail out those doors
for the first time, heading for deep space, three hundred people
on board, everything working, the Mind quite happy, and you think,
I helped build that.
The fact a machine could have done it faster doesn’t alter the fact
that it was you who actually did it."
"Hmm," he said.
"Well, you may 'hmm' as you wish," the woman said,
approaching a translucent hologram of the half-completed ship,
where a few other construction workers were standing,
pointing inside the model and talking.
"But have you ever been gliding or swum underwater?"
"Yes," he agreed.
The woman shrugged. "Yet birds fly better than we do, and fish
swim better. Do we stop gliding or swimming because of this?"
He smiled. "I suppose not."
"You suppose correctly," the woman said.
"And why?" She looked at him, grinning. "Because it’s fun."
> excels at the part of the job that was my favorite (writing code)
AI excelling beyond a human at writing code is not an objective truth.
I would say even without AI, a manager telling a developer what to code or how to code, will never be better than that engineer themselves being the driver and the developer.
For me personally, quality code has a strong human component. Only I know what's best for the code, based on:
1. my deep understanding of the long history of the organization and of its mission and concepts
2. my intricate personal relationships with the users and the other developers
3. all of my hundreds of unique, personal experiences troubleshooting problems, which are always a combination of code and interpersonal relationships, the open source community, and opposing stakeholders.
Until AI becomes AGI (and that's a big if) and you can train it on 43 years of experience, it will just be an approximation of a true person's holistic engineering pedigree. It will always be asymptotic to true competence. It will always be second-best, even with a human driver.
I have friends that enjoy knitting and sewing. Those are great pastimes made commercially unviable for most because of great leaps in logistics and automation. But they are absolutely valid and enjoyable hobbies.
And the skills they provide are transferable. Patience, attention to detail, staying on task.
At that point it’s probably worth taking a more mainstream hobby like music or dancing, you’ll have a much bigger community of passionate folks. I suspect that will happen to programmers as a hobby folks too, they might discover they like playing music instruments instead.
Not to mention your friends still enjoy receiving custom made knitted stuff as presents, or your neighbors if you see someone struggling. Or donate it somewhere, or put it in a display shelf to be reminded what you can actually do. Maybe in the future, programs and custom made hardware might be the same :)
> Maybe in the future, programs and custom made hardware might be the same :)
One step closer to the matrix. Send the white rabbit over, I've got your floppies ready to go. No ai, and developed without an active internet connection, so you can listen to music safe in the knowledge that your music isn't listening to you.
It's just tough to hear a mainstream "impactful" line of work be compared to knitting and sewing. Say what you want but people thought that their work was necessary to the advancement of society, or at least something bigger than themselves. Whether you agree with them or not, reduction of the job to a hobby takes that away from them.
There's also a tension in your comment. Are we prioritizing "advancing society" or programmer feelings? If it's the former, then it doesn't matter if AI is better than humans, whatever moves us forward faster is better. If it's the latter, then we're talking about holding back progress for the sake of a programmer's feelings.
For the record, I'm a lifelong programmer who loves AI because now I can experiment with new ideas at a scale I could not before. I feel zero loss. In fact using AI brings back the magic of technology that we lost in the 90s with the "business-ifying" of tech.
I'm not saying we're advancing anything. I'm just saying that before programmer feelings used to be aligned with advancing society: fun work (for some people) was useful work. Now that's starting to be decoupled and it sucks when your choices are hate your job and contribute or do fun stuff and don't contribute. Granted this is the dilemma most people have faced, but for people who like their work as it is right now it sucks.
I'm a programmer and I'm having more fun than ever thanks to AI. I do think a lot of programmers are pretty conservative by nature, so maybe that's the group that's struggling more. If I had to wager, I think the camp is split between those who spent a long time in university vs self-taught hacker types (I'm the latter).
I'll take open models any day of the week. We're almost there but they're not quite good enough yet. Re: fun, don't you have a backlog of incredibly weird experiments and projects you'd like to build? I did and now I'm building them, it's very fun (and educational).
The people who spent a long time in university are actually better off than the self taught hacker types and this has mostly to do with the fact that so many people go through university without learning how to program.
University is not easy, so how did they skip the easy part? How did they get good enough grades everywhere else to pass?
Turns out: They must have done something that has nothing to do with programming but it is still computer science. What was it? What did they learn???
> Granted this is the dilemma most people have faced
I don't agree. There are many creative and useful fields: engineering, law, finance. And others that don't require as much creativity but may be enjoyable due to social factors: medicine, teaching, emergency services.
The dilemma that people have faced is that they don't want to do earnest work, but still need money to live. There are enough types of work that are useful and stimulating in different ways that everyone should be able to find one, so long as they are given opportunities as children (we do a decent job of this in the developed world). At least, that was the contract. Many fields, not just software, are facing existential crises.
I guess when I said what you quoted, I meant for most people (developed and developing world) throughout history, or at least as long as "work" has existed.
I find this comment fascinating for a few reasons. The first is that it has the underlying assumption that all "progress" is good (regardless of what we're progressing towards). From my perspective, it seems like most of the progress being made is towards bringing about a sort of techno-feudalism and environmental collapse. Maybe a little bit less progress would actually benefit humanity?
The second thing revolves around this:
> In fact using AI brings back the magic of technology that we lost in the 90s with the "business-ifying" of tech.
This is the complete opposite of my experience. AI seems like the logical conclusion of the business-ifying of tech. Get rid of all those pesky nerds, and the MBAs can just create whatever grift pops into their head.
> From my perspective, it seems like most of the progress being made is towards bringing about a sort of techno-feudalism and environmental collapse. Maybe a little bit less progress would actually benefit humanity?
I'd argue you're conflating two independent axis: technological progress and societal inequality.
AI is useful today and is getting more powerful and efficient. This is obviously great for humanity! It's enabling discoveries that were previously out of reach, and allowing us to accomplish the same amount of work with less effort.
AI is also generating massive amounts of money for the wealthy and powerful and increasing inequality. This is not great for humanity!
I do think it's important to keep in mind that this centralization of power and wealth is not caused by AI directly: the pattern exists whenever there's a desirable but scarce resource. The same argument that AI is increasing inequality can reasonably be made about concrete, steel, and steam engines. The problems caused by AI are really human problems at their core. It's possible AI could even help us fix them! (I'm not at all hopeful.)
LLMs haven't enabled any discoveries that were previously out of reach. If you disagree, name one?
>AI is useful today and is getting more powerful and efficient
No, it's getting less efficient at an exponential rate without proportional performance gains
>do think it's important to keep in mind that this centralization of power and wealth is not caused by AI directly: the pattern exists whenever there's a desirable but scarce resource. The same argument that AI is increasing inequality can reasonably be made about concrete, steel, and steam engines.
The goal of these LLM companies is to funnel all economic activity through their models. LLMs only exist because of how they strip mined human work without consent or compensation. I.e. LLMs do directly cause the centralization of wealth and power by how they were designed and how they are meant to be used.
> LLMs haven't enabled any discoveries that were previously out of reach. If you disagree, name one?
While not an LLM, protein folding is now often done via transformer.
> No, it's getting less efficient at an exponential rate without proportional performance gains
AI is expanding in both directions. Sure, it's eating the world with its compute demand and the bitter lesson bites, but we're also seeing drastically improved efficiencies employed alongside. The cost per task is dropping fast.
> The goal of these LLM companies is to funnel all economic activity through their models.
That's what all companies want, AI or otherwise.
No arguments here on the ethics of AI companies though.
Programming is the art of automation, and is being taken to new levels. It'd be quite damming to the field if a substantial % of the world's output over many decade hadn't brought us here, tbh.
Were builders crying when the Hagia Sophia was built? Projects come and go, nothing is forever.
interstingly i think what you are talking about is pretty apt, but not the way you are framing it. knitting and sewing were impactful things in their time; clothing has been something humans need for a long time and more skilled craftsmen were able to produce hgiher quality garments. new techniques increased durability, wearability, comfort, etc. machines commoditized that and made it mich faster to produce garments, so turned these into hobbies and the odd artisan fpcused and made new improvements, but not as often as before.
programming will likely be the same. fewer people who care to be craftsmen, more hobbyist, a few artisans who really dig deep and make new patterns (we still dont see AI sampling out of distribution).
>reduction of the job to a hobby takes that away from them.
this is your personal feeling. they never had anything in the first place, just that the market was valuing a certain skill for what it could output. the perspective you are posting is purely a construct.
Sure it is a construct (it's how I feel at least).
I didn't mean to say sewing and knitting weren't impactful - on the contrary, they are analogous to programming: they were impactful but now they're basically hobbies. Maybe there are more developments left in the clothing making process - not sure.
I think you're undervaluing the skill and technology behind clothing, as well as the impact of clothing. But that's probably a reflection of how technology has allowed us to take it for granted.
it's like how I can still get a typewriter and use it for the experience, or like how I could write ORM scaffolding by hand or just let the code gen write it, I see AI as nothing more than a keyboard extension, the amount of freedom you give it depends on the instruction/prompt
There's still a big ecosystem of people selling homemade sewn/knit products online, selling at craft fairs, selling sewing patterns, running classes, etc.
Even in something as heavily automated and mass produced as clothing, there's still a market for smaller scale, higher quality products.
As a sewist, when I sew things, it's also because the mass market doesn't make the exact thing I want. The software mass markets, both proprietary and open source ones, has also failed many consumers. The developers in Adobe tells me to pay $20/mo for PS, while the developers in GIMP takes 20 years to implement proper text outline. What is the point of enjoying how the software is made if people don't get to use it for what they want?
These skills are all less valuable today, btw, along with roughly anything that would have allowed someone to thrive economically pre-2025 in spite of occasional accusations of high-functioning autism.
The issue is not that programming is getting automated. The issue is that the only skills that retain value are in low supply among programmers - and every other career you've considered retraining for.
The problem is that AI is creating a race to the bottom. We are forced to use AI tools and burn tokens because leadership don't have a clue. And even if you're doing complex work, people will assume it's just a prompt...
I enjoy writing complex programs because I get appreciation for my work. I don't really care that much, but I'm sure we humans love being appreciated for what we do.
This AI shit has degraded most of what I enjoy. Artists, musicians, and writers are drowning in a sea of slop. Mathematics is looking like it won't have a human future. Programming is something you can't get paid for.
Same.. and now everything can be done SO FAST it's just stupid.
I used to love taking the time to whiteboard my project, write the code in peace (with maybe a background show). Go through the steps of structuring, debugging, testing.
Now I can write a prompt to basically do what I do instantly, and now I'm the manager and tester of something I didn't truly even make.
So sure the upside is I can do more, and I basically have to use it because I can't even keep up with what people want me to do without it. And it also seems so silly to go back.
We were lucky for decades to get paid to do what we enjoy. Virtually no other field gets this. Maybe artists but note that very few artists make even a living from their work. So yes it sucks but as they say, "don't cry because it's over, smile because it happened".
That's the economy. It's not like all our problems are solved or everything that needs building is finished, even in very AI-forward tech companies. We are busier and working harder and longer hours than ever, and we've been hiring thousands of people in India.
I suspect that many of the loudest voices pushing back against AI coding tools never practiced programming as a hobby or saw it as especially fun, and the fact their job is changing won't make them start now.
I think just about everyone in the industry appreciated hand-typed programming as a job, because it was a pleasant enough craft to practice and for a solid couple of decades programmers were treated extremely well in comparison to the vast majority of other careers out there.
Today, juggling a bunch of agents to do what you want them to do, figuring out how to verify their work, remediating problems, and spreading yourself thin cognitively—all the while looking over your shoulder at an increasingly precarious job market—is notably less pleasant for a lot of people. It sounds and feels a lot more like an engineering or product management role—something a lot of programmers would never in a million years have signed up to do voluntarily.
I'm probably an 8 out of 10 on the cooked scale—I'm juggling a suite of ten iOS apps and haven't written or read a line of code in over a year—but I get where these people are coming from. I think the suggestion that they can reclaim what they've lost in this new job they've been loudly expressing distaste for by adopting a new unpaid hobby will ring hollow to them.
Wouldn't people who thought programming was fun be even more upset that they can't do it (a certain way they used to) as a job anymore? That they were never working a day in their life before and now have to work every day?
Yes. This is much closer to how many people are experiencing it and why the OP's suggestion of letting people practice it as an unpaid hobby doesn't address their issue.
It feels like spending your teens practicing football, playing varsity level in high school, getting really good at the college level, finally recruited into the NFL as a pro football player (dream came true!!) and suddenly, the team owner says, "No more football playing for you. You get the same salary but you're managing robots who play the football. Deal with it."
Computers can destroy Magnus. Yet he still plays and nobody gets to tell him or anyone else they have to stop.
If people stop being interested in human-played chess, then yes he's out of luck. But that could happen with or without computers. Same for football and anything else.
Of course the analogy works. It’s an analogy, not an exact duplicate of the situation.
The important parallels of the analogy are that:
* Magnus dedicated his whole life to chess, just as many programmers have done with programming.
* Magnus enjoys the craft of chess, the human elements of creativity and cleverness, just as many artisan programmers do.
* Both prefer to be hands on and don’t enjoy being relegated to managerial / supervisory roles, and would probably move on to something else if they could no longer do the part they love.
* Many outsiders do not have any idea how much time and dedication it takes to reach the upper echelons of a skill like that, and struggle to empathize with those who have done so, and perhaps even feel Schadenfreude to see these folks “taken down a peg.”
Mathematicians have recently been lumped into this category of knowledge workers at risk. Many are grieving gratuitously and suffering a lot of flippancy from Schadenfreude-feeling onlookers.
No, it doesn't work, because both analogies (basketball and chess) are entertainment/games. Magnus isn't being paid to produce the best possible chess moves, he's being paid to be the best human chess player and beat other humans. A basketball player is similar. So the act of playing chess, and the act of playing basketball, is what provides the (entertainment) value.
A computer programmer, on the other hand, is paid to produce something. They aren't entertaining anyone (except themselves). The act of producing code doesn't matter to the person that wants a binary. The binary is what matters. So if a computer can out-program a human, then the human is in trouble. So it's different.
But, nothing is stopping a human from programming for entertainment purposes. And nothing is stopping humans from being entertained by watching other humans program, maybe competitively. Like chess, or basketball. Or from getting together and solving programming problems. Then it doesn't matter that a computer can run circles around them, and nobody can tell them to stop doing it.
I get it, though. Younger people in tech should be worried.
This is not what I said and you're being rude. I'm saying that I don't understand the grief being expressed for the loss of a hobby. Because as I explained, computers can't invalidate hobbies.
> Sounds like I got suckered into arguing with an AI yet again.
Nope. I'm a human and didn't use AI for anything I wrote. But thanks?
The point is that what was once a life's calling is being reduced to a hobby. All these activities were more than hobbies, for example for people like myself solving significant math problems all day or at least for a significant portion of my day was my life's purpose. I felt like I was contributing to something greater than myself. Math can still be done but that feeling is gone now and with it is much of the motivation to do it in the first place.
The “entertainment value to the customer” is the product being produced by the professional game player. That’s what the team owners pay them for. I don’t see much of a difference. The programmer is being paid to use his skill at programming to create business value. The sportsball player is being paid to use his sports skill to create business value.
We've commoditized programming because it often gives an output that can lead to many beneficial things. But that does not mean that programming itself isn't entertainment.
Making art is often considered entertainment, and it can also be commoditized. The difference between entertainment and work is often thin, dictated by personal enjoyment, not how one is paid. I can be paid for something I find very entertaining. I can also be paid for something I find dull.
If the thing I find entertaining disappears, it's reasonable for me to feel grief. For something I enjoyed is no longer available to me, at least not in full.
There's a community around programming, much like chess and basketball. Seeing that community deteriorate into solely programming for profit, and using AI to do so, is tragic. Programming has been very focused on profit for a long time, but now even those who think it fun are debating on whether or not to offload their work to a machine that does all the work for them.
Speaking for myself. Not really. The problem that I had with enjoying programming was that many of my coworkers who didn't enjoy programming have some opinions that were frustrating to need to entertain.
Now all I have to do is make sure they're not peddling an actively bad idea and if not I can just tell the bot to incorporate their dubious ideas.
In the past it was all, I'm not familiar with this construct that solves our problem perfectly, why don't we cludge adapter pattern in here again.
If they haven't worked a day in their life before, then who has built all the software products we use today and from which LLMs are endlessly benefitting?
>I suspect that many of the loudest voices pushing back against AI coding tools never practiced programming as a hobby or saw it as especially fun, and the fact their job is changing won't make them start now.
This makes no sense. Everybody I know, including me, who hates the big picture of what AI has done loves programming immensely and does it as a hobby as well as a job. The people at my job who are most positive about AI are those who didn't particularly enjoy programming. You're applying Occam's Razor backwards.
I would agree with this. Programming/coding has been a portion of my job as an Ops engineer for last few years, but it was the part I liked the most. So I’ve been pretty heavily against AI use.
However there are other folks in my company who, with very little programming knowledge, have produced viable, workable solutions for our company to use. All projects that were in my list of TODOs that I hadn’t gotten around to.
It was in that moment that I realized I’m very quickly going to be surpassed if I don’t adapt and move quicker with AI solutions. Since then my output has increased a bunch.
I’m not the most stoked on it, but as with all companies, results are far more important than the journey. So long as it works and is at least mostly maintainable, leadership does not really care how we got there.
But all of this was forcing my hand because people without the foundational skill set were creating equivalent output to my artisan, bespoke solutions. Writing is on the wall.
I think you're entitled to your opinion, but want to point out that there's 2 different axes on the graph here. You are unhappy about LLMs' affect on your working conditions, but also concede it's inevitable and the labs are at least somewhat sincere in their predictions of it displacing a significant amount of the value in the software engineering market.
It's not necessarily a given that the person you're replying to agrees on the latter, and therefore they might not be stoked on you starting off by agreeing with them but going on to make an argument that supports the companies responsible for the very thing they're complaining about.
I think there's a contradiction in you stating that management thinks LLM-generated output is "maintainable". They went to business school, they don't know what that means. It's your job to explain to them the consequences of shipping code faster than the people whose job it is to understand it can do so.
> It's your job to explain to them the consequences of shipping code faster than the people whose job it is to understand it can do so.
The problem we’re collectively running into is that nobody knows if this actually poses a problem anymore. We’ll find out in 5-10 years, but businesses and clients can’t wait that long.
As a side note, I don’t think it’s fair to blame the “management” for setting expectations around AI use. They’re driven by market forces. If competitors are producing slop and customers are gobbling it up, then from a business perspective it _is_ a mistake to abstain from AI use.
What I’m finding quite fascinating is that we’re in a weird spot right now where we’re both expected to produce slop AND understand it / be responsible for it. Seems unreasonable.
I think the most important takeaway is that it takes time for these problems to manifest. If you start with a particularly coherent codebase it will take years for slop code to degrade it to a point where engineers are noticeably less productive.
I didn't downvote you, but if I had to guess why you were downvoted it's because it's very obvious to many working software engineers that decoherence is a real problem. In many ways that's what this OP was all about; not just loss of joy in programming for its own sake but loss of the ability to design deterministic, fully-understood systems that won't require constant off-the-books maintenance eating into your free time.
I don't know if it's just me, but I find that with a disciplined approach (or even an undisciplined one for some personal projects), LLM generated code is quite "maintainable" nowadays for the sorts of things that matter: adding new functionality, fixing the odd bug (usually a missed edge case, or behavior I didn't think-through well enough). They're also secure enough for me not to be overly concerned.
Granted, I stick to frontier models and avoid cheaper ones, but if it quacks like a duck...
If you're working on a CRUD/web app, absolutely. There's so much data for web apps out there that LLMs now produce better web app code than most of my coworkers from 2022 or so, since it follows consistent style rules. Web app code all looks the same, anyway.
I've never particularly enjoyed programming on its own merits, but also think AI is crap. Its mainly because AI clearly accelerates the parts of programming that I've found tend to lead to software projects falling apart, ie not understanding the code
The last 20 years of experience have told me that that's the #1 variable that correlates with the death of a project, so I'm in no hurry to speed that along. People have been trying to find some way to skip that difficult step of actually understanding what's happened since programming was invented, and its always a bad idea - because that process of internalising a projects structure is the single most valuable thing for a project. Creating a project structure that can be internalised by human beings - and then doing that - is software engineering
There is some truth to that, thus i wonder, if they will not solve that by creating architectural harnesses. Like look at repo x. Condense its coding style and architecture into a prompt. Remember that prompt everytime you create a new module.
It only remains to explain this to the management... That does not work with code, engages with AI only in the form of chatbots, but demands everything to be "AI native" and won't be there to deal with the consequences.
I feel the opposite, pushing keys was never the enjoyable part, deciding what types go in what places, uninteresting, switching between editor and terminal, boring at best, trying to find translate errors to SO search queries and back again, maybe possible, dealing with under-documented dependencies, frustrating.
My goal has always been to create something and use it. Ai still requires we do the intellectual side of programming, that interesting side of programming. I get to do more of that now that while the bot handles that annoying stuff. I'm a the happiest programmer I have ever been. I spend more time writing human these days, because the bots are bad at that.
>This makes no sense. Everybody I know, including me, who hates the big picture of what AI has done loves programming immensely and does it as a hobby as well as a job. The people at my job who are most positive about AI are those who didn't particularly enjoy programming. You're applying Occam's Razor backwards.
I think I know people in both camps who feel both ways. I'd say it's all just noise at this point. What's clear is some people like it (for now) and some people don't (and probably won't ever). I've gone from liking it to not over the course of the last year. I used to love programming/the craft, and then got burnt out of it, so I loved it because it made building things easier again after I stopped caring about that, and then it made work hellish when everyone basically stopped caring about the craft at all.
I feel for everyone in this situation. I don't think anyone is wrong to mourn the undeniably massive shift happening to something that has been such a big part of our lives, and I don't think anyone is wrong to be excited. I say this as someone who has worked in ML for coming up on 10 years now and is more towards the latter camp, where I love working with agents, and still spend plenty of time writing code by hand and studying. I don't think it's fair to pretend people are being unreasonable or are just "missing the big picture" for being unhappy.
But I also have found that where you work and what AI usage is like in your company is a pretty big deciding factor. Your comment, which I agree with, made me think about this. I consult a decent amount, and I've worked with companies where the engineering culture has adopted AI in a way I describe as "ugly." I should find a better word. But what I mean is that it's sort of like the attitude is "okay, now we can ship more tickets, so everyone take on 8x the workload". And they do a lot of "best practice" kind of stuff around using coding agents--every new anthropic article gets posted to slack, and there's always a new CI process for using agents--but they're fundamentally just trying to do the same work as before but kinda sorta faster. It's a weirdly miserable experience.
On the other hand, I've worked with companies where the engineering culture almost feels exuberant. I think a consistent difference is that in these companies, it feels like the engineers now have permission to try wildly ambitious things. It's not always optimally productive in the sense that more things get thrown away or end up just not really working out, but overall they move significantly faster and do more interesting work. Commonly, it's things they wanted to do before, but simply weren't realistic with their constraints around capacity. Or it's an idea like "this part of the codebase has always been a performance nightmare, but it's also a beast to untangle. I wonder if I could setup an autoresearch loop and let it run all day, even just to harvest some ideas?"
But yeah, when everyone sort of gives up and/or are forced to move an ungodly number of smaller tickets, it is a miserable time.
I definitely know what you mean about the difference in engineering culture. I'm definitely the kind to get excited by it and try out all manner of experiments and really have my head in the game on the new paradigm. My current team just isn't in that same wavelength. It's a lot more practical focus on shipping, and fair enough. It's a bit of an impedence mismatch for sure. Still, I have fun playing around with stuff in the evenings at home, but now at work I'm just trying to do bog-standard engineering and forget about the AI side of it to the degree that I can.
Hard agree. My anecdotal experience: me and my friend group have math/coding contest national level background. We loved coding. All of us now devastated and consider retirement/different profession
People who was there for the benefits (no offense, pretty valid reason) are happy now
I declined manager position multiple times, now I am forced to it
And no, I can't code without agents as a hobby anymore. As a rational person, I feel like this is just useless nowdays. Huge part of the fulfilment that it was the most optimal way to solve the problem. Now I feel like I do useless work
> The people at my job who are most positive about AI are those who didn't particularly enjoy programming.
At least two people I've worked with have become "AI Influencers" on social media. They were two of the least capable programmers I had ever worked with. Both super nice people, and I enjoyed their company, but both generated absolutely terrible code.
I think there's generally something to people who were sidelined by their lack of skill feeling empowered by AI.
Just yesterday I got a fully LLM reply from a coworker about an issue.
I've never been the biggest hobbyist programmer. My hobby programming was a means to an end for the actual hobby project I wanted to use. I'm more of a "maker" than any one kind of engineer. That being said, it was still fun at times and those personal projects literally got me jobs and allowed me to bypass live programming tests because I was able to show I knew how to program as I explained every design decision.
LLMs sorta fucked me over. My strongest skills aren't raw programming knowledge or fact regurgitation. Personal projects were how I showed potential employers my actual end-to-end problem solving. The projects were never polished, they didn't have every error condition formally handled in a logging framework. But they organized ideas into data and made things happen.
All that is worthless now unless employers want to see my pre-war steel collection on old repos. I'm defeated. Using agents for personal projects feels like playing a game with all the cheat codes turned on. It stops being a game, it's pointless. Me telling the machine to make something doesn't prove anything.
I still want to remain housed into my senior years and this tech has possibly kicked that out from under me.
> It sounds and feels a lot more like an engineering or product management role—something a lot of programmers would never in a million years have signed up to do voluntarily.
True, but dealing with agents is still easier than dealing with people (at least for introverted personalities). That being said, you still need to deal with other people's army of agents.
I enjoy puzzles as much as the next guy. But I can just pick up a crossword or sudoku or something and get that mental stimulation elsewhere. The thing I always found vexing about programming was that the puzzle-solving was a prerequisite to getting anything done. Now that the two are decoupled, I'm a lot happier.
Not a few engineers are actually doing stuff like planning the cooling for buildings, construct things in CAD or draw plans. There might be some people involved afterwards that actually put those plans into reality and then it becomes a supervisor job similar to what you describe. BUT, thats only part of the job, not the core.
I see it similar to programming, before it was kind of a craft, many crafts are not paid very well, I know that woodworking can be very fulfilling but Ikea exists and noone pays you 1000€ for a simple table. But programming was a well paid craft because people are bad at thinking in general and not everyone can do the thinking good. Now we externalize the thinking and hope it will just work somehow...
But the point is..that is a different job. Others already said it, suddenly you are a manager and not a craftsperson anymore. No more tinkering.
About the hobby stuff, I hate that the unannoited ones slop there way into _my_ hobby space, pollute everything with one shots that get abandoned in a few hours and drown everything with their half baked brain childs without polish that all look and feel so same-ish, the "map of something something" projects I have seen that were missing any kind of error validation or check of completness is astonishing.
> I suspect that many of the loudest voices pushing back against AI coding tools never practiced programming as a hobby or saw it as especially fun, and the fact their job is changing won't make them start now.
As many people here and on Reddit can attest, I am one of these loud voices.
I actually see programming as all three: a hobby, fun and a profession. And AI itself is... neither. It's not enjoyable because you have to understand everything anyway if you want to do a decent job, no matter if for hobby or for work. The only thing AI can do is save you time when you have no starting point at all - i.e. reconnaissance in pentesting.
Are you serious? I LOVE programming as a hobby. In fact is IS my hobby. And I HATE AI, hate it with a passion. The world was so much more interesting when people were doing it by hand. Now it's ugly.
> It sounds and feels a lot more like an engineering or product management role—something a lot of programmers would never in a million years have signed up to do voluntarily
Now that you phrase it that way.. I think I'm very glad to have transitioned from 'programmer' or 'dev' to 'big-tech software engineer'. The jobs are very different. I had no idea what I was in for. SWE is much less fun. All meetings and alignments and docs. But I think that's what my job would have been forced to become with AI anyway, so at least it prepared me for this new world.
Some of us enjoy mentally modelling everything to see opportunities to iterate as each cog turns in their head. Often without being in front of a device, letting sub conscious figuring out difficult problems. AI has its place but so does that process.
If that sounds like you, then don't be discouraged to keep thinking.
Obviously entrusting an entity to do everything is different from using a dependency or library, but aren’t LLMs for work just a natural progression of this? Programming is about automation. Programmers have been automating and outsourcing their work since the concept of
SDKs were invented. Or even since high-level languages were invented. The industry was already doing this, and doing it excessively, long before Claude. left-pad happened a decade ago and we’ve learned nothing.
Not really. Even with left-pad there was still some skilled work required. At this point my wife could vibe code an app or a website and it would be actually quite good - not the slop of 6 months ago.
Writing left-pad isn't the issue, the issue is that projects were importing a one-liner package and blindly using it, which requires no skill.
>Second, even if the package’s logic is correct, I can’t help but be amazed by the fact that developers are taking on dependencies for single line functions that they should be able to write with their eyes closed. In my opinion, if you cannot write a left-pad, is-positive-integer, or isArray function in 5 minutes flat (including the time you spend Googling), then you don’t actually know how to code. Hell, any of these would make a great code screening interview question to determine whether or not a candidate can code.
> Finally, stringing APIs together and calling it programming doesn’t make it programming. It’s some crazy form of dependency hacking that involves the cloud, over-engineering things, and complexity far beyond what’s actually needed.
Obviously LLMs make the situation even more dire, but I think people are overlooking that the industry was automating itself out of programming as a craft even before them. It's just the automation involved reuse of the work of other humans.
Sorry but there's no way that assembling libraries - even stupid ones like left-pad (which I'll note was an extreme example) - is in no way comparable to just saying "Claude make this program for me", which is pretty much where we are now. My wife can do the latter but not the former.
This isn't just "a bit more automation". It's "now we don't need lift operators".
Of course, prompting an LLM to one-shot an app is a quantum leap in comparison, but c'mon, the whole direction of the field is towards automation, abstraction, convenience. It's always trended letting the programmer write less code. Have to deal with less menial tasks. So LLMs are just the latest manifestation of that motivation. And, perhaps in a decade or two, something will come along to replace LLMs, and this conversation will happen again, except even with more desperation and despair.
But while your example exists, I don't believe that's how the industry works, yet. They're not just allowing laymen to one-shot their production software. Certainly some teams may be trying that, and some indie devs have made good money already selling vibecoded software. But I don't think the technology is good enough to auto-generate professional code at scale, without human oversight. Give it another eighteen months.
Detractors say it causes needless suffering, accelerates global warming, bad for you in excess. Some say it simply doesn't taste nice, and are met with "Have you tried wagyu-7.1, it's much better than sirloin-flash-3".
It's really not like that. Sorry to disappoint you but it's more like cars were invented, and someone says "have you tried the Ford Model T, it's much better than a steam carriage".
Or if you prefer "have you tried the new iPhone, it's much better than the HP iPAQ".
Its a bit interesting how the very fact that some technology has different grades of performance can escape people.
Its something I've noticed in many folks. You'd say something like "well sure, did you try using a different model" and you will be met with screeching. Its almost as if they've shut off facts and made up their minds.
An example: a year back I wanted to show that AI doesn't hallucinate that much.
Here's like a 20 comment long conversation with a person who believed that ChatGPT hallucinates easily. It took them a long time to prove it - they used toy examples and it did't work because they'd never used the thinking version. Then they had to resort to esoteric examples in the benchmark which still didn't work.
I'm a little in the middle, I was always really good at programming, it just clicked for me. As a physicist, I could see that I was OK at the math and physics but excelled at the programming piece. At the same time, I never really cared much about the art of programming itself. It still was a means to an end. I know that does make it so I can't totally empathize with engineers who just love the art of programming and the existential fear they might be feeling with the looming AI takeover.
Then you're just a stranger in a land of nerds, only knowing about the history of computing you would know that coding isn't only a means to create something but it is everything. Would the Michelangelo statue have been 3D printed hold the same value? The problem is that programming got popular because it paid good salaries
Yeah I don't understand that attitude. There are so many people on twitter who say things like that or "X industry is cooked" etc. All pushing that narrative hard. Why are there so many people pushing it so hard like that? Are they all under the influence of the big labs and it's all guerilla marketing? Are they bots? Are there a lot of people who have drunk the kool aid or suffering AI psychosis? Normally when new tools come out this doesn't happen. I can only think of a handful of times where I've experienced this in the past where someone has pushed something on me really hard (Ruby and Rust spring to mind).
To be fair though, the vast, vast majority of software developers out there are working on CRUD and line of business apps that a) change all the time, b) are probably irrelevant a year down the line, and c) just aren't really sophisticated at all. There was never really a need to have finely crafted, artisanal code in these code bases, but it was something developers enjoyed doing, so they did it. Yes, sure, maintainability, yada yada. For most code out there, that never mattered anyway.
So AI is kind of just bringing that to its logical conclusion.
Do we have that middle ground? It seems like something many people are desperately looking for, but are not finding right now. I think we’re all open to your solomonic wisdom, if only you’d share it with us instead of making snarky remarks.
Most of the times someone writes something like that I find they are mostly talking to themselves.
Great to hear that the author found peace somehow, though I would advise not to worry too much about what you consume out there if it affects you to the point of needing justification.
Same not sure for the same reasons, to me it all sounds like a big marketing campaign for AI coding. Everywhere I read I see comments bragging about that the quality of these products output is better than us or it is usually taken for granted. Not my experience so far and I'm on the edge of AI usage
There was a comment I saw in a thread today. In this thread someone expressed concern over blackboxes and people not understanding them.
A reply I saw bluntly stated their opinion which could be summed up as "Good, ignorance is bliss".
Whatever your thoufhts are on LLMs, it has never been more apparent that HN is not a place for hackers.
I'm sure some SV tech fag who was little more than a CRUD monkey a couple years ago will show up any moment now with their own definition of "hacker" as tends to be the case on GrifterNews.
> I still program on the weekends because it is fun
Fully agree. It's still (or even more than ever) more entertaining than watching Netflix. Why should I care if other people use agents to do the programming?
The frustrating part isn't so much being asked to use AI to program—it makes sense in a lot of cases—it's watching people blindly use AI in a way that you know, being experienced, will end in disaster. And knowing that any argument will be dismissed under the pretense of obsolete thinking.
I've resolved to just building stuff for my own businesses and myself for fun. If an opportunity or enlightened soul comes along that wants to benefit from my knowledge, great. But I think it's best to just let people feel the full brunt of what's coming. Sometimes the only way to learn the lesson is to stick the fork in the socket.
>it's watching people blindly use AI in a way that you know, being experienced, will end in disaster. And knowing that any argument will be dismissed under the pretense of obsolete thinking.
Experiencing this led me to feel burnout. It's horrible. I now value thinking and the ability to think much more. Not that I didn't before, but I mean I have a whole new appreciation for it and a whole new disdain for Claude slop where someone could have otherwise done a better job.
This basically correct but there is some nuance. First of all, people working on unoriginal or very small scope stuff, such as app clones or ad hoc analyses in notebooks, will never hit that wall. They'll never even know the wall is there. This is a bit annoying, because they then generalize their experience with toy projects and come away with the mistaken belief that anyone could make anything if only given enough tokens. This is most problematic when they are in leadership positions.
However, the real value is in complex, non-trivial, and above all novel code. This has been true for a long time, and it's just as true in open source as it is in industry or academia. Here, agents are far less useful. In fact, the only people I've seen have any kind of success using agents on these kinds of "serious" projects are people who are already serious programmers. They already know how to design and test stuff, they already know how to supervise junior developers, they already have an intuition for which aspects of the app the can safely delegate to a less experienced developer. So they're able to use agents as a "better autocomplete" in their own code, or use it to flesh out the non-essential and more routine aspects of the project, or even just delegate the agents to the role of code reviewer.
Here's the thing though. The less experienced ones, the people who don't already have those skills, they do get burned when they "stick the fork in the socket," as you put it, or in other words when they try to use agents to take on projects they don't fully understand... but they don't learn anything! The ones I've seen, anyway. Relying on agents insulates you from the code, which also prevents you from building a mental model of what's going on under the hood, which in turn makes it impossible to learn anything at all. They just keep yelling at the agents, telling them to fix stuff even though each fix introduces yet another regression in an endless loop... they don't know how to dig themselves out of the hole they've dug for themselves. And after all, what is the solution? To spend five years down in the code mines, paying your dues as a junior dev, reading and typing and thinking about mere code for the requisite 10,000 hours? Nobody has the time or inclination for that.
Agents, in other words, break the learning loop. You can't learn to program, and therefore can't learn to design or architect software, not really, if you're using agents. I am unfortunately coming to the sad conclusion that the last programmer has already been born; if you didn't have 5+ years of experience and a pretty good handle on the more advanced aspects of programming (such as software architecture, design patterns, optimization, debugging, etc.) before agentic programming became a thing, then I don't see how you ever will.
It'll be like COBOL programmers - the industry will shrink from the top-down, with the first to go being the new college graduates looking to enter the workforce. Inexperienced devs have always been given less interesting projects, but they stuck it out because they were learning valuable skills, building experience, making the economically rational decision to invest in their own career. But anyone entering the market today will never get out of that trench - anyone that actually needs a real programmer for a serious project will hire someone with decades of experience - there will be plenty of supply. And the projects on offer for junior devs won't teach them how to program, only to supervise agents. As a knock-on effect, smart young people already see the writing on the wall and are pivoting to other majors, which just reinforces the trend. This might cause problems in a few decades when all the real programmers start to retire, but when has the market ever considered long term consequences?
> Yes, at work, where you are paid to produce value for a business, the economics of purely hand-written code have changed
I'd love to genuinely love these blog posts but they are all the same! They're all like "now that machine are better than us $xyzargument". I'm so pissed off when I read so blantant lies like AI "writing code better than me" maybe better than you! But in general it's just a big lie.
> the economics of purely hand-written code have changed
who are you to know? I still hand code and produce a lot of value because I do care and polish everything I do. There are some pieces of the software that I wrote that I never touched again, they work perfectly, no errors no customer complaints
It depends on your definition of better. Some businesses value speed above almost all else. Especially early-stage businesses where time to market and test ideas is paramount. I think in those cases, LLMs are better than probably any human.
On the other hand, if you mean tight, performant, simple, well-factored code, then yeah. LLMs simply aren't there, and may never be. On the other hand, maybe that matters less than we thought it did. Time will tell.
> Especially early-stage businesses where time to market and test ideas is paramount
To me it's just crazy of how dumb it is to test an idea with a vibecoded shitty app. I'm not saying it should be perfect from day zero, let's pretend we are talking about cakes, how crazy it would sound you will tell people to try and judge your new idea of a cake you don't even know how it taste or what's inside?
I agree. You shouldn't test an idea with a vibecoded shitty app. That doesn't mean you shouldn't test it with an LLM-generated, refined, well-tested (by human users) app. I don't think it's a boolean. I think it's a continuum.
Those companies, and those people within otherwise decent companies, have existed since long before the advent of LLMs. Avoiding their influence is a career skill as important as the code itself. And I'm not saying you can just avoid them for your whole career, they're everywhere and are being supercharged by LLMs.
It's disappointing that the prevailing sentiment now is that architecture is less important than we thought. Software is always going to be getting more important, and all of it should be well thought out. If these people had these tools in 2004 they might've made thousands of fragmented buggy social media startups and actually slowed progress on that front.
How is this different than testing an idea with a janky, rushed, handcoded protype or MVP? That's how a lot of product development has always been done.
Well there’s at least two reasons people might not buy your cake: they don’t want cake, or the cake is bad.
For the first group, it doesn’t matter how good your cake is. If you spend a year perfecting it, you won’t sell them any more cakes. And for many products, this group is basically 100% of customers.
I think their major issue is actually naming. I don't think they have issues making well factored code because you can tell them how to factor it. You can even tell them how to name it. But that takes time. People want the functionality now without having to wait. So the well factored and well named stuff is a functoin of how much time you want to spend giving overview. I don't think you're going to have many cases where its faster to type each thing yourself. You can describe the details in a prompt which can be referenced as you review.
The issue is nobody is spending time in review because the stuff coming out "just works". So if no one is reading the code, what is the big deal with naming? The big deal is we need to know what our later sessions are talking about and we need linguistic separation of concerns.
A higher level form of programming, if you will. Linguistic precision seems to be where the specificity can be lost or won. The naming will reflect that with some degradation. But humans will also degrade the spirit of the specifications as well in many cases!
Great point and I think it highlights the issue with talking about this topic because the terminology is so black and white, "AI made it" or "I made it". But the truth is more like I got on a bicycle that I did not build and rode a marathon through treacherous land that I have not been through before, but I have been through other treacherous land before and I had some scar tissue that helped me through this time. Jobs was on the money when he referred to the computer as being like a bicycle. ("for the mind" ... to finish the quote)
Well I’m sorry you feel that way. I don’t see “great point” used often but in terms of its agreeableness I can see how one could take it that way. (In fact I had a feeling someone would say that, but it was a concise wording for what I intended so I left it in).
I see it as a multiplier (the bicycle metaphor) and yes certainly getting across town on a bicycle or across the country on a plane is less cumbersome on your health than walking every where.
I don’t ride motorcycles, I rode a scooter for a long time doing a delivery job as a younger man and the speed and instability (of two wheeled motorized vehicles) along with lack of visibility to others creates a dangerous combination.
And yes I can see the corollaries there to AI usage especially in these days when the excitement of the acceleration is being found seductive by many and without much visibility into the longer term consequences.
> The issue is nobody is spending time in review because the stuff coming out "just works".
People used to be able to write code that was horrible by all the standard metrics but still "just worked" (or at least appeared to). From the stories I've heard, it would get into production all the time, too. But if it's a bigger problem now, then necessarily we know how to fix it, so.
I guess the excitement of the new toy is just leading to some irresponsibility.
I don‘t know that we know how to fix it so well after the fact, but rather we know how to avoid it which is by writing understandable code the first time round.
But we don’t always do that and then things get rewritten and broken etc
Welcome to life in software dev 2026, not so different from 1996
I don't know that it's a lie. The code that I can get Claude Sonnet to produce is pretty much what I would and that's two iterations behind. We've vibecoded scripts recently that genuinely are better than what I would have written and I don't think I'm terrible.
There are a lot of mediocre programmers aren't there, I wouldn't be surprised if on average Claude is doing better than what exists. It's not a "lie", anyways, just a point of view.
> We've vibecoded scripts recently that genuinely are better than what I would have written and I don't think I'm terrible.
I'm ok with vibecoding scripts, I love programming the machine and making the end user happy not dealing with bash! I don't judge scripts quality as long as they are tools in the workflow, I dont care about that, unless they will somehow be part of the production. The name Script implies that, there is architecture and Claude Opus sucks at architecture
This. I hate that Claude Code is taking the fun part of my job, but the reality is that Claude Code Opus 5 can one-shot a script in 15 seconds that would have taken me 45 minutes to write.
I don't begrudge the business for making the prudent choice of making me use AI, I just sort of lament that what I spent years enjoying and getting good at was so utterly derivative that it was perfect for the LLMs. It would be insulting if it weren't just a fact.
Every time you don’t write the script you lose a little of the ability to do so.
Just like if you don’t read you will lose the ability to.
These skills are not innate to us and can be forgotten over time.
I wonder if it will get to the point where people don’t even know what to ask for. Or how to judge if what the machine produces could have been better. Or if it does what they think it does.
Not wholly. It’s under study. I don’t mean in the sense that you can’t read words anymore. I mean in the sense that comprehension and attention go down.
Yes, sadly, coding is a skill that can and (before long) will be forgotten by all but a niche dedicated few. It will go the way of assembly language, smith-ing, weaving, bread-baking, meat- / fish-gutting, hunting, remembering phone numbers, driving manual (and, before long, driving in human mode), hand-writing, horse-riding, wood-chopping... you get the picture. The price of progress!
...are you sure that Claude Sonnet writes code that you'd write?
For me, it writes code that mostly works, which is great. It insists on adding some of the least readable comments I've seen. It also likes to pad out the code to about double or triple what I would write.
It doesn't seem to have that little voice that says "hey, this approach sucks" so it needs the harness or the prompter to tell it to go back and refactor. Even Fable and Opus struggle with this.
> It doesn't seem to have that little voice that says "hey, this approach sucks" so it needs the harness or the prompter to tell it to go back and refactor.
Well, yes. For humans, the "this approach sucks" voice is still very separable from the thought process that just figures out the syntax and algorithms etc. The important thing is whether the overall system has this feedback and that it's having the desired effect. Better harnesses are progress in the same way that more capable models are.
And yeah, there's a wide range in the standards that programmers hold, in practice. Plus they don't at all agree on what's "better" than which.
Regarding the "this approach sucks" voice, there's an important difference between humans and AI which I've noticed and which I didn't explain well.
The general human approach to programming at most companies I worked at prior to 2024 was to head off in vaguely the right direction and refine the approach as we went. There might be a design and a plan, but at some point we just winged it and fixed our work after the fact.
This seems to be the worst way of getting an AI to do anything. They do better if they can one-shot an implementation from a clear spec then only need to fix bugs. Better models can handle a less clear spec but they still can't really iterate in a major way. To have an AI model build a bigger thing, you one-shot increasingly detailed specs and decompose them into smaller ones.
I can only speculate, but my guess is that this limit is pretty fundamental, yeah. However capable the models get, they can't read your mind. The kind of iteration you have in mind is difficult because they don't have a basis to iterate upon; to the extent that they could pretend, that gets into what people have been calling "taste".
And even if they somehow got better taste, it wouldn't be your taste. If that were acceptable, why bother prompting anything in the first place? At that point the system might as well choose its own projects entirely. But I don't think humans would want to live in that world; even if not paperclip-maximized, we'd eventually become an underclass.
That's assuming basically unlimited capacity for the models to get more powerful. If there are limits (and I hope there are!) then it'd be even more certain that projects continue to require human guidance.
But even despite that, I hope I don't sound AI-psychotic if I suggest: the models are capable of at least assisting with that spec-writing and decomposition.
The industry is in a desperate race to throw away what little quality it had (could you imagine if other fields of engineering had catastrophic failures at the rate software does?) in favour of "look at all the lines of code I generated today but don't actually read them please". It's always been a fad-heavy field, but this time it's really gone fully off the deep end. The whole "it's better than me" is just re-asserted endlessly. Nobody even pretends to care about evidence. I wish the boosters would go ahead and "leave me behind" already, I've been waiting years for it to happen now. The zone has never been more flooded, I think.
> I wish the boosters would go ahead and "leave me behind" already, I've been waiting years for it to happen now.
I completely agree. The fact that the software from places that are very vocally pro-AI (microsoft, salesforce, etc...) is still so full of bugs is really contrary evidence.
> The whole "it's better than me" is just re-asserted endlessly. Nobody even pretends to care about evidence.
The evidence that the AI tool is better than oneself? I don't know about you, but I think I'm a pretty good judge of whether Claude Code is better than I am, having seen my own work and Claude's firsthand.
Exactly. Someone on HN put it really succinctly earlier in the year: "LLMs look like they are good at things you are not good at."
I think this goes a long way to explain why there is such a huge difference in opinion about the quality of AI-written code: Some people say LLMs write code far better than they could, and others say the code is always terrible and requires a great deal of fixing before it's acceptable. How can two different developers say those very different things about the same product?
It might be a tough pill to swallow, but if you think your LLM is producing better code than you could produce... well, I might have some bad news for you about your own skills.
In my case I am more than capable of producing better code than the LLM, but I have never been the fastest on my team. My development speed was acceptable because the quality and robustness of the output was worth the time that thoughtfulness took. I was also able to improve the comprehensibility and robustness of my colleagues' code during code reviews.
But in the new environment of LLM-assisted coding, the LLM produces acceptable code so much faster than I can write it. Just taking time to understand and retool the LLM's output means an order of magnitude lower productivity and it doesn't feel like that's acceptable in this era.
Folks I know are landing 10+ PRs a day and nobody is reviewing anymore, and these are not companies that would have been skipping code review 2 years ago. The whole culture is turning against taking the time to comprehend and reason about our own work.
Here's the thing, it doesn't have to produce code better than me.
There is a wide range of code that will do the same job. 15s vs 2h for code 60% as good? 10 minutes for code 80% as good? 10 minutes today and 10 minutes tomorrow to fix something that is 30% as good?
A lot of people are going to draw that line in a different place and now that people can make that trade off you will quickly find they will and it probably wont fall where you would like it to.
At the end of the day a business pays you to produce value that will earn them money / customers. They don't pay you to write good code. Sometimes those two align and both sides are happy but they don't and often wont.
Eh, my understanding is that the models are supposed to have guardrails against that. But also, getting better results generally requires refinement, a critical eye, and the skill to write out what's wrong. I'd guess these skills aren't common among that audience. But much more so, it likely defeats the purpose. The point is to be lazy (that's at least part of why people are willing to spend money on it).
On the other hand, I'd also expect the most degenerate among us to not care about the uncanny details. So who knows, really.
If you can't get it to output good code I'm afraid it's probably a skill or prompting issue. There are programmers that are almost certainly better than you that are using LLMs regularly to produce good code. Do you think the Linus is a bad programmer?
Linus has motive behind his position on AI. It's his operating system that's at the center of the AI revolution. The Linux Foundation is involved with several AI projects and has a lot of members that are all-in with AI.
He is not a programmer anymore. He admitted that he mostly reviews, tests and gives suggestion now. He still does a little coding, but usually small patch that hard to say if he is good or bad programmer.
I like to think I hold very high standards; I find that I can fairly easily get the LLM most of the way to where I want the code to be, but there's definitely a Pareto cutoff.
I do feel that Codex and Claude are both bad at something I'm good at: writing documentation. But it doesn't matter, because no human reads it. I'm spending a lot of time rewriting its output just for it to be piped back in again.
And that's just sorting out my self-image. Convincing someone who controls the pursestrings that I'm adding value was hard already.
Thank you! It's made me lose a lot of respect for my field how credulous everyone is to any argument that starts out with a fig leaf of pretending to be on your side.
"The AI labs are scared of their own models, surely they must be credible sources on how powerful those models are"
"They are asking the government for regulation, they must be even more responsible than the average company"
"They are claiming to be a software engineer who automated their job with AI, surely I have to buy their course"
There are many boilerplate pieces of software that an LLM can generate faster than a human, but nearly half a decade into the LLM era there is no evidence that it's leading to the apocalyptic end of software organizations. And that's all with the subsidy in place that is now disappearing. Everyone online is lying for money, and that money is coming out of the same investor-subsidized marketing budgets.
Although "lies" might be too strong a word, I agree with you that the the AI-is-the-future people haven't done nearly enough to prove their claims. But that's nothing new in our industry, really.
I would recommend "The Leprechauns of Software Engineering" by Laurent Bossavit for an insightful and entertaining overview of just how much of software "engineering" is based on "truths" that we don't question.
You’re forgetting the “economics” part of the argument. Let’s assume that you are actually producing better code than say Opus 5.5. You’re probably slower, you very likely have less “stamina,” and you probably cost more than 200x more per month than a $20 subscription.
You’d have to be pretty unbelievably stellar to produce code 200x better than Opus. I can get a lot of business value out of not-great $20/mo coding. The people writing our checks do not care about the same things we care about.
And I say this as someone who is very upset about what this stuff is doing to my career and hobby.
I don't get this. I work in a part of the industry where "passes 80% of the test suite" is worth approximately $0/mo, which is the best opus can do without a human hand holding it.
But I read the OP's argument to mean that your responsibility as an engineer is now to understand the tasks an agent can accelerate. And in many cases that might be the task that you previously enjoyed, like working out some tricky variable length encoding serialization scheme, or tracking down some nefarious race condition by stepping through code and reasoning. (The eureka moment is a big dopamine hit for me, dunno about you.)
If you can't get Opus 5.5 to pass more than 80% of your test suite, then you're probably holding it wrong. Not "holding it wrong" as in Apple's "we designed a defective product and try to put the blame on you", but as in *"this is an incredibly sophisticated tool which takes skill and practice to use to its fullest extent".
LLMs, even in a good harness, are deceptively simple, but in reality rather complex tools. They're good at grasping intent, but they can't read your mind, they might not have access to the full context of your business, and they require learning how to use them properly.
No, not "hand holding". If you insist on anthropomorphizing the AI, "guiding it on a high level" would be more fitting. In a team of 1 architect and 5 software developers, Opus 5.5 + Claude Code replaces the 5 developers, not the architect. You don't have to "hold its hand" anymore than you have to hold the hand of a competent software developer — though you will have to figure out a workflow that suits you personally, including configuring Claude Code to follow that workflow (through skills, for example) so you can work ergonomically with it.
I have never worked on a team with one thinker and 5 minions, so, again, more evidence I work in a different part of the industry than wherever your line of thought originates.
Your error lies in thinking Opus 5.5 is equivalent to a "minion". The "5 software developers" in my example were meant to stand in for "reasonably competent, above-average developers", and not that the architect is the only one on the team who can "think".
Of course, it may well be the case that you are working in an elite sector of the industry where even the bottom rungs are filled by the top 20% of software developers. Possible, but unlikely, so I'm still leaning towards "you're holding it wrong".
Opus has to be 200x worse for them to sell that subscription. If their AI is as good as they claim, they should charge way more, like one dollar less per month than a human developer costs.
>I'd love to genuinely love these blog posts but they are all the same! They're all like "now that machine are better than us $xyzargument". I'm so pissed off when I read so blantant lies like AI "writing code better than me" maybe better than you! But in general it's just a big lie.
Yeah and this is what pisses me off about the software factory deniers. Without the software factory to guide the agent and reject its code, how exactly is it supposed to do better than an experienced developer who already internalized the company workflows? All the Agent sees is its context window, it's not a fully formed human that can learn the workflow over time.
If I guide the Agent file by file and function by function, then review and reject the code, the code quality becomes reasonably good, but when I let it do whatever it wants, the code quality reaches all time lows and it just compounds on itself.
I also noticed that the code quality heavily depends on the underlying software ecosystem with Javascript often having the worst code quality because it presumably got trained on minified code as well.
Agent harnesses do not solve software development in a hands off way. You still have to baby sit them. Every single time I use an auto review based workflow (aka vibe coding) it just doesn't work.
I am in the minority I have the feeling. I started as hobby software developer when I was a child, then I graduated in bachelors ICT (as it was named then) and have now worked for 25 years in a mix of public sector and companies. Simply said I got tired of coding the last few years. I did not see myself coding till my retirement, but was not sure how I would get out of coding.
Then AI came a few years ago and that was interesting as a tool to play around with and do some simple small code (snippets) generation or as a buddy to discuss problems and find solutions. Meanwhile I was getting more and more bored with implementing stuff that I did already so many times at previous jobs. Same principles, different code base.
Last year I started at a new job and was first hesitant, but I decided to go 100% AI with doing a rewrite of an existing application that will be phased out. So actually I could develop from scratch with the help of AI. First I copy/pasted code, then this year I found out about AI agents.
Never in my life have I been so productive in creating features/functionality while the code that AI generates doesn't differ that much from what I would have written myself. But then in a few prompts and 10 times as fast. Including unit tests. Which at many places that I worked were more often not a thing, then that they were.
Looking back at my 'career' it also made me realize that I do like "tech", but don't like coding as much as other software developers, I guess. I really like AI to make functionality and things working in a very short timespan. I enjoy 'AI-coding' much than I'd have thought. So many things that you would not even try because it would cost too much effort, time or money. Now I can test within a morning a few different features to see which one fits best and works best.
Never have I been such a happy 'developer', I don't miss the coding part. And AI will only improve, so personally I don't see a bright future for a lot of software developers to be coding in the traditional sense.
> AI generates doesn't differ that much from what I would have written myself
How do you know that? Do you read AI code? I heard (DHH, Uncle Bob, my own co-workers) that reading the code is no longer an economically viable activity. And if you are not reading the code, how would you know what it looks like.
Oh, I read the code. I even write it, at the risk of losing credibility at work.
I claim that AI "writes 80% of my code". But realistically I am in the hot path of my agent ("cd repo", "claude", "do X", "fix Y", "no not like that"), and that's essentially equivalent to writing by hand. I am afraid others will soon see through my charade.
Personally I still try to read or scan the generated code and look at the implementation and many design patterns are recognizable and are almost always exactly as how it should be done. I must admit that there are sometimes small parts of implementations that look inefficient or 'weird'. I remember that one time it wanted to store data in some sort of array type in a database instead of a separate table. Asking why AI did that it also agreed that it was a what tricky/shaky implementation. I didn't trust it and with a single prompt the code is rewritten into a perfectly fine separate table implementation.
That's the reason why I still try to inspect the generated code to see how my problem was solved and if it reads plausible. There are also moments that I am surprised by the implemented solutions which look very elegant and "smarter" than what I could have come up. These are also the moments I learn from AI. I can even ask AI why it did things certain ways and it gives a nice summary of arguments. But I am often also critical towards AI if I am in doubt, resulting sometimes in a sort of discussion and sometimes I can convince AI (with more context) that the solution it created should take another route. It really feels like working with a really fast co-worker. And on top of that, I am better in writing up my thoughts to discuss things instead of talking with someone. Doesn't mean I don't discuss in real life with co-workers, but it is so easy to use AI for it too.
518 comments
[ 0.18 ms ] story [ 32.4 ms ] threadOpen source development has become buried in an avalanche of AI generated code as well so the character of a lot of hobby programming in that domain has changed drastically.
If you’re a The Art of Computer Programming-reading, code golf-competing, blogging neat elegant hacks kind of programmer, were you really getting that experience coding at work?
Change comes to every profession, and it's never universally helpful or productive. I'm very much in the middle of the road on the AI booster/skeptic continuum, but I know it's not going away.
And getting to "enjoy" the work was always just a perk. Where in labor law in any country throughout history guarantees enjoyment of the work? Surprisingly few people throughout history have been able to say they enjoyed their work. It's always been a trade of time for money, everything else is table stakes. I'm finding new ways to find enjoyment both in the job and out of it, and if that time-for-money trade stops being worth it, I'll figure out something else.
0: https://www.newyorker.com/magazine/2018/11/12/why-doctors-ha...
How would you reverse this trend, I dont know, but AI basically puts us in the same boat as the doctors in that article.
And that's OK.
These are talkers who have never produced anything, taken everything from open source while looking down on OSS authors. They one-shot a note taking app and feel they have done something culturally significant.
AI in software is the greatest farce since NFTs and attracts the same people. All web software got worse, nothing new is being written and everyone just talks about their precious conversations with the clankers like commenting on a football game.
The dumbest people always win temporarily.
Perhaps everything that you said is 100% correct, but people are easily swayed by things that “sound right” being overly aggressive opens your position (in people’s monkey brains) to “enlightened centrists” that leave the door open for bolder opposition.
Many people have been genuinely helped or feel like they have been genuinely helped by LLMs in coding, they aren’t going to change their opinions based on fiery rhetoric and accusatory tones. Much like I can’t force myself to believe in Santa Claus regardless of how aggressively my nephews claim he is real.
it's hard to be swayed into skeptic camp if you see how it works and it really solves those hard problems that took hours. Genie is out of the bottle
Furthermore, it becomes practically impossible to argue for doing things the hard way for all of the solved problems, which were and are necessary for being able to learn the craft deeply. At the very least, deeply "enough" to advance the state of the art.
In short: 1) Humans are necessary. 2) Knowledge and experience are necessary. 3) Humans need time and practice to gain knowledge and experience.
When one side argues that one or all of the above points are false, I don't believe them.
Realistically, we are seeing people even in the most pro-AI groups begrudgingly concede #1 and #2, while expecting #3 to just sort itself out.
really? i feel like the overwhelming sentiment in every comment section is staunchly anti-ai. it's far more common to see people lamenting their job prospects or complaining about having to review mountains of slop PRs.
In my opinion, the whole anti-LLM coding thing is purely ego preservation. And I get that, who the hell want's to believe their high paying job is going away? To say using LLMs cannot achieve results bigger than 'one-shotting a note taking app' is really just not true.
Comparing it to NFTs is absolutely wild also. NFTs literally were mass psychosis and had no value whatsoever. You're saying someone can't build software, even just for themselves, or otherwise, that doesn't have any value at all? How could that be true?
You are on an online forum geared towards programmers, using the world wide web created by a programmer. The concept of a video game did not exist before they were being programmed.
You think the world just appeared as it is and everyone only does things for money? Never because, unlike people who use every shortcut possible, they actually care about the details?
LOL as if AI was a not a mass psychosis ... "AI have 10% of kill us all"
If you go with agentic route, randos with money to spare on latest models, beat you. If you go LLM-free, AI corpos say "all the better for us! More free untainted dataset!"
Humans are so lazy and obsessed with cost-cutting, efficiency, and "line go up" that we'll willingly walk ourselves into such a bleak inhuman future.
They are happy to have their ads-maximized agreed-to-all Cloudflare/Anthropic/OpenAI/Google/Apple/Microsoft accounts so they can one-click deploy that "worker" with all the UI gears spinning automatically, because who needs to know how to configure a web server anyway? :^) And wow, have you heard that Cloudflare offers an operating system now??
If you had to larp as an anti AI person just to say the stupidest thing possible, this is what you would say.
My problem with AI is that it now excels at the part of the job that was my favorite (writing code) and it leaves me to do the part that I suck at (being a manager). In 18 months or so my job satisfaction has gone from pretty high to abysmal.
I worry about burnout.
Then the next day after this brain haemorrhage inducing experience I just waste 2h remotely on gathering mental capacity to simply know what next I need to do and next 2 says I repeat until weekend.
Simply because my boss is developer in love with LLM and machine gunning 20 PR daily.
Not to mentioned PR with more code than I can reasonable can understand and say anything about it in a project I didn't touch.
So, currently I can only say I hate my life. I'm looking for an idea which pays similar but is as mindless as humanly possible
Do they know they are, essentially, expending you?
It seems like your boss is doing work that doesn’t actually add value without you to review the code. But maybe the company isn’t set up to fairly evaluate the fact that review provides more value than creating code, with a workflow like your boss’s?
I don't about the OP's specific situation, but there are definitely managers and management consultants that consciously try "expend" their employees.
Jeff Bezos is a famous example. IIRC, to avoid "complacency," he wants constant turnover of employees. Hire new people, burn them out while they're new and enthusiastic, then replace them. That's why it's occasionally leaked that Amazon consider such-and-such employment market mined out (e.g. everyone's either worked for them or is wise to their ways), so they have recruit further afield.
https://jacobin.com/2023/03/automation-union-busting-product...
I went from being a passionate programmer spending all my free time working on code, to be unable to write a single line for even my most loved personal projects for 6 months. Fully awake, in the middle of the day, I would boot up my computer and within 10 minutes my body would put itself to sleep. I'd wake up hours later with my laptop barely past the login screen. It took me 6 months before I could do personal projects again. 1 year before I could do professional work again. 2 years before I could juggle both. I've fully recovered (or at least I did right before LLMs hit the scene), but it takes a lot of time and I lost a lot of money not working for that year. If I could go back in time, I would have quit that job earlier because it is far easier finding jobs as a passionate developer rather than a recovering burnout.
And even if it does end up being good code, if it's something non-trivial, having a reasonably deep understanding of what it's actually doing is quite hard simply from reviewing the code.
I keep giving AI-assisted programming a serious try, but time and time I land in the exact same place: reading the code is more exhausting than writing the code, _and_ I end up with a far worse understanding of what is being built than if I had just written it myself. The fact that LLMs tend to write _way_ too much code certainly doesn't help.
And what makes me sad is that, if you follow this line of inquiry, you eventually realize that capital isn't challenging the value of humans _writing_ the code, but rather the value of humans _understanding_ the code. Because, as I've experienced many times when talking to those how have completely bought in on the hype, the answer is always to just "let the AI handle that." AI writing too much code? Have it summarize it for you. Your understanding of the code is worse than if you wrote it yourself? If it works, who cares!
And yes, I think we are well on our way to becoming Markdown jockeys.
Ever since AI reached a certain level of quality, I haven't been able to stop (vibe) coding. I've discovered that I really enjoy designing and building products, more than tinkering; occasionally, I come across algorithms or other technical details that I find interesting, and I switch to manual mode.
At some point in my career, I just got tired of having to read tons of documentation for fragmented libraries and frameworks, and AI is a "solution" to that.
Cynically, this could be seen as plumbing, but there's an art to designing efficient and usable products, that I personally enjoy very much.
People enjoy different things. Its not strange. Treating it as strange is kinda insulting, so I'd recommend that you rethink how you phrase this.
Yes! Which means that the argument frequently seen here of "AI is terrible because everyone hates using it" is not valid.
It's all projection to make themselves feel better about their difficulty adapting to change.
My problem is it doesn't even do that. LLMs still aren't good at writing code (despite the hype) and you have to review everything they produce, which takes just as long as writing it myself would. But the industry is falling all over itself to use them to write code, doing a worse job than humans would. It's extremely frustrating.
From what you're always commenting, I could believe that you were exclusively using Haiku 4.5, which would explain how the tool produces only garbage while the rest of the world pretends AI was great.
Just out of curiosity, do you use like Opus 5.5 in Claude Code, and you still find the results inadequate?
Something has to give; the constant onslaught of slop and workflows changing every few months just isn't sustainable. Either we slow the fuck down and figure out what we're even doing here, or the AI improves so much that we're all out of a job.
This is ok. You will process the change, and you'll come out stronger for having done so. It takes time to decide what you want to do in the world, and even more time to decide how to react to the world changing and moving away from your planned path. And it takes even more time to actually see if the world is moving in a way incompatible with your plans.
It will be tempting to cycle into negative predictions and negative self-talk. This will not help, and will not be productive.
You, ryandvm, built a career around something you love, and that is amazing. You can choose to do it again, to find something you love, to be stalwart about how you work and who you work with, or any number of other options. The thing that nobody can change is your agency.
I struggle with the nature of work every day, and have for months. It has caused me to re-evaluate what I work on, why I do it, how I structure my finances, where my kids go to school, literally everything. Still the ship turns, as it must.
Reach out if you want to chat.
* I made the classic mistake of assuming the present was permanent. It never is anyway, small changes always ruined the present. The accumulation of small changes over my career has changed the way I work probably as much/more than AI has changed it this year, if you can believe that.
* I discussed the family budget with my wife, and made a plan to halve our income requirements. It will require a race in the short term, but allows a coast later on. This is key for me - almost all my tension regarding this phase is about the future earning potential of our household. If that goes away, who cares how work changes?
* I made the other classic mistake of not remembering how much I have changed and adapted over time. The breadth of things I have worked on is pretty remarkable, and the things I have learned to do are crazy. While things are changing fast now, they aren't changing quite as negatively or immediately as some claimed they would. There's room to adapt.
* I stopped assuming I had to work the way people said I should. Aside from CEO mandates, there's a ton of wiggle room in how you produce code. I read changesets and make manual adjustments to the pattern, etc. I write my own parts when I want to. I specify a design that I want rather than asking for whatever the AI wants. I get to own everything but the typing that way, and my RSI is improving dramatically b/c of it.
* I use AI to set up programming challenges when I feel disconnected. It sets up a tangible problem, and acts like a coach only when I need it. I use this to preserve the feeling of sharpness. Earlier this year I challenged myself to write, in C, from man pages only, a "ping" and "traceroute" clone. I did so on a flight back from a work trip.
* And I started reaching out to friends to hang out more. It's hard to doom about AI when your friends and you are having a beer and playing Container or Captain Sonar. Life has always been what's "out there" not what radiates from your monitor.
Derek Thompson, in "I Know Why the Mathematicians Weep": "If intelligence becomes less uniquely human, we may come to value more highly what remains uniquely human: our capacity to relate and to care for one another".
Funnily my bot was called Cladius van der Bot, nice name ;)
People could learn things on their own but still pay teachers
Trainers, coaches, therapists… yes.
even the job of freelance engineer is customer relation. how good are you at extracting product requirements? understanding a clients workflows? these take relational skills and are often what freelancers are paid for, beyond the ability to produce code.
I doubt most people will be developing a rapport with an AI agent trying to sell you that contract at work, or helping you when you're irate and have escalated a problem a few times.
All of those except for therapy are awful career fields. Musicians, actors, artists (outside the tip top of the field) are all the butt of jokes about how broke they are. Social work chases people out once they realize how powerless they are for such poor pay.
Therapist might make out well for a bit, but it seems like a loaded term these days .
If you're aiming for the top, sure. There are plenty of people who make a living doing stuff like that and are happy because they enjoy it. You're unlikely to be one of those people if you move to an area where there's an extreme over-supply and try to "make it big", or try to be "pure" to the art and eschew commercial options. e.g. An actor in LA, the starving artist in a loft in NY.
> Therapist might make out well for a bit, but it seems like a loaded term these days .
Loaded how? I was thinking of psychologists, licensed mental health professionals, physical, occupational, and speech therapists.
Typing the actual lines of code was fun, sometimes, but it was never the goal. I feel like I'm still solving the same problems.
Even if one were to continue doing everything by hand, it will still be harder compared to others who aren't. As in, companies will come to expect more/faster work than you are capable of, so it still affects us old farts that don't want to change.
We'll find something else, sure, or just retire... but not everyone is/was ready to.
I'm not sad on a personal level per se, I can still be positive and look to the future, but I lament what we increasingly no longer have.
I too have lost something I loved, I gained it after years of relentlessly focusing on maximizing what I loved and minimizing what cost me energy. And now the world flips on me.
The good news? I got pretty good at consciously finding the energy givers and drainers and optimizing for the givers, so I'll just keep doing that. I'll either start enjoying micro-managing LLMs, or find something else entirely. Or find that the LLMs need less and less micromanaging, or start enjoying whatever they need next year. (Point an army of agents in a direction?)
I came from the molecular biolab with Linux as a hobby, then when to software dev and infra. What's next ? Idk, I still love biology, keeping eyes and ears open...
I've always taken the view that it was luck - lucky that my interests aligned with the market, lucky the market put value in what I am interested in doing, and lucky that I was given the chance.
I was never in it for the money, I was always in it for the enjoyment, and I will remain in it for the same thing.
The market for my knowledge and skills might come and go, there might have been a brief moment in history when someone like me could make a living off these skills, but like so many other creative careers that time was always going to be fleeting
No amount of processing will move someone from ordering a steak to just enjoying their given peanut butter sandwich instead. They ordered the steak, and now the restaurant is telling us all too bad there's no steak anymore - eat your damn sandwich.
To adjust the metaphor a bit.. the cows have revolted and we can no longer get any steak.
If writing code via AI is an order of magnitude faster there won't be companies hiring people who want to code by hand. If you try to start a company by doing so then you'll lose to competitors without that restriction.
The choices left are: be homeless, change careers completely or suck it up and do something you don't like after being career-catfished
When you think through the 'change careers' path it also sucks and might not even be viable for many
The comment you're referring to aims to be helpful and empathetic, but the fundamental problem with it is that it essentially dismisses the very real problems people are raising. It tells you that the change is inevitable and that you shouldn't and/or can't fight it. It tells you that what you're dealing with are your own feelings and, although they're "valid" (in the sense that it's understandable and human to have them), you need to work on overcoming them or finding a way around them. And if you can't, then you should either move out of the way of the progress or you'll remain unhappy.
It's an incredibly frustrating way of being talked to, one that I've seen managers employ quite a lot. I call it the "management aikido", because it takes your own impetus and turns it back on you, by reframing whatever issue you brought up to put the onus on you. Most of the managers I've seen use this technique believe they're doing a good thing by "empowering" you to deal with problems. What it does, in fact, is discourage you from raising issues in the first place.
On top of having the same "management aikido" vibe, @jvanderbot's comment ends up being "socially insensitive", for lack of a better way of describing it.
Over the course of history, we've seen great shifts -- typically tied to technological evolution -- that caused significant suffering to a group of people but (arguably) ended up being a net positive for our society. And when we look back, we inevitably feel a sense of empathy for those people and we recognize that what happened to them sucked and we express the hope that maybe next time we could do better as a society.
And then it happens again, like it's happening now, and we not only fail to do better as a society, but we talk like trying to do better is unreasonable.
That, in essence, is what makes @jvanderbolt's comment sound "religious" to you: the unquestioned assumption that what is happening is supposed to be happening and that it's you who has a problem that you have to work on.
I'm definitely stealing "management aikido" by the way, thanks for that. :)
I don't feel like it's patronizing to remind people about achievements they've accomplished and that they're still a capable person. Nothing we say to each other here on HN should be taken as "This is how you must live you life" but more like "Here's is something I hope helps you, if not, no worries", as we hopefully assume good faith out of each other.
I love solving problems, programming was always about that for me. The sucky part was always "I know the solution and written it out, now I need to use a text editor to actually make it work like that", and the fun part was always figuring out how to solve the problem, making sure it actually does solve it in real-life, and once it's been implemented, iterating on it until it gives people some sort of pleasure to use.
Most of the other passionate programmers I've met in life have thought about programming the same way, all of us favored approaches with less code, not more, because typing sucks and is slow, and more code means more bugs. I haven't met any of you "I love typing special characters into a computer" people in real-life, and I'm wondering where I could, because clearly it isn't just you who feel like this, seems there is a somewhat large section of programmers who enjoyed the typing experience more than the problem solving part of programming.
What's always been clear, is that people get into programming for different reasons, and some of those reasons might no longer lead to the lucky life of working with what you love and get paid handsomely for it, which sounds terribly sucky and scary, and I do feel for the people that now have to face this.
I've also never really experienced that professionally. I started as a professional programmer in 2012 and I've always been pushed to do some form of "agile" where you figure stuff out while you work because building detailed specs upfront is too costly and time consuming
So now I'm being told it's fine to pivot to these AI tools and it's awesome, you just plan upfront and execute. But I genuinely don't think I have those skills. My workflow has always been "have a rough idea where we're going and fill in the blanks as we go"
So now I'm trying to adjust to this new work, planning upfront with more detail and I'm struggling.
Coding is how I used to work through problems. I was really good at it. Now I feel like a fraud. Just a hacker, not a software engineer. I don't know how to get to where I need to be, and I'm really unsure I even have it in me to learn and entirely new way of being
I went from feeling on top of my life, my career on a great trajectory, and really confident I could build anything that anyone needed, to feeling completely worthless
And there's such little empathy out there for anyone going through it like me. AI is here now loser, keep up, get on board or get replaced.
If you read this far thanks. And if anyone has any advice for me, how to go from "thinking through problems by coding your way through them" to "doing the thinking upfront" then really, I'm all ears.
I think that this really needs to be internalized by a lot of people. I really appreciated when my boss sat our team down and said "Yeah this is the new normal, and I know its not what you signed up for. Its fine to feel sad that your job is changing in such a dramatic way." No, it didn't help anyone actually adjust, but sometimes you just need someone to empathize and tell you its OK to feel bad about this when everyone around you is excited.
> If you read this far thanks. And if anyone has any advice for me, how to go from "thinking through problems by coding your way through them" to "doing the thinking upfront" then really, I'm all ears.
I'm trying to replace it with "thinking through with AI" to limited success. A lot of people recommend having the AI "plan" the thing and review the plan, which is OK. I've also had it do a prototype and poked around a bit, summarized the learnings, and then did it again in an iterative manner without rough edges I've found. Whether you are given the space to do that is up to your org -- its hard sell to "do it again" but I've found that the second attempt leads to better results since I can prompt my way around the problems. Sometimes this also leads to a "Not even wrong" moment where I found out that I framed the problem wrong and the AI dutifully carried out doing the wrong thing.
Comments such as "now that economically 'traditional coding' no longer works" get over it. Or a recent one "get over the anthropocentric view of intelligence AI is here now".
These views (and many more besides) are short sighted, I do think anthropocentrically. The economy is a proxy for the lives of real people, the economy matters only because it deals with real people. People who feel, have agency, who live and die.
Two super AI's talking to each other maybe the most intelligent things in the universe but they don't matter.
Humans matter, human lives matter, human feelings matter.
One-shotting without review is very quick but has its downsides: burnout, less understanding of the code base, less interacton with your colleagues, you miss the poor choices the LLM made so the code base might rot.
Writing everything manually is much slower and in some circumstances not what you want anyway.
Is there a middleground where you get significant parts of the productivity/speed increase but retain the most important parts of the manual process? I have found I preferred the copilot style of working, but the interface was not really optimized.
Instead of spewing out code at 20x speed can we settle for 5x and remain in control? That would be my ideal and more sustainable, right?
Unfortunately, I don't have a lot of first-person advice for you; as I say in that essay, I was also a bottom-up programmer for a long time and was deeply skeptical of top-down approaches. What changed my opinion and my practice was just doing things for long enough that I started having a sense of where things were going to go before I did the exploratory work to find it out, and eventually the exploratory work was unnecessary and I could create a basically-correct top-down design early on. I don't know how you build that experience without actually getting in the reps (which has some implications for future generations of developers, if we have any).
That said, my second-person advice would be to use the agents for the design part of things, too. "Here's the problem, how would you design it?" and then have a conversation with them, push back on the things that seem wrong, imagine reviewing this as a finished design from a coworker and seeing where it looks right and where it doesn't. Ask why it designed things a certain way; suggest alternatives for it to evaluate and think about what it's weighing. Basically borrow the robot's experience to augment and build up yours.
Who is forcing you to use AI?
It is an amazing time to be alive, scary and uncertain, but amazing nonetheless :)
Outside of LinkedIn Lunatics and social media engagement bait, I agree that nobody is saying you can't enjoy programming.
Of course we can all still enjoy programming. I still do some personal projects manually. It's fun and it also keeps my skills sharp.
It also makes you realize how much non-optimal stuff slips into entirely AI written codebases if you're not paying attention to the code. I can push on Opus 5.5 or Astra all day and get output that works, but when I get into the code and start processing it myself I ask a lot of questions that result in "You're absolutely right!" responses followed by having the LLM clean up what it did.
The sad part is knowing that the golden window or programming job enjoyment has closed. We had a relatively small window where software jobs paid extremely well to do work we enjoyed. My happiest career memories were working with small teams at startups where we would sit in a room together and be focused on coding for hours.
I'm sad for a second reason: The years before AI took over, my jobs had been trending away from that focused coding time for other reasons. The trend in the industry was to stuff companies full of roles like product owner, scrum master, cross-functional project managers, and other roles. These are highly valuable in small quantities with the right people in place, but at some point it got out of control. Most of the companies I was with felt like a constant circus of meetings and debates over what to build, with little time left for that focused coding work anyway. I still cringe when I remember a meeting with a CTO who explained his target ratios for engineers to designers to product managers to engineering managers to project managers where we realized software engineers were becoming a small minority on the software engineering teams. I disliked it at the time, but now I'm extra sad that my last window of possible focused paid-to-code time was spent mostly in meetings and Slack debates.
When you do that, a team can just go. Instead, in many places there's all this bullshit about "Application X is built by Team A, but Team B and C will do the QA." Or "Team D is in charge of Applications Y, Z, and Q, so they can't get to your feature request for Q this PI, because they've got high-priority work for Y and Z." So now you've self-blocked all work on Q just by screwing up your org chart. And we haven't talked about the fact that Team A is in the US, Team B is in Europe, and Teams C and D are in India, so now let's add time zone nonsense to all of the above stupidity.
You don't need as many of those ancillary roles if people involved actually understand Lean and how to run an SDLC, but in my experience, there are too many devs who don't care about anything other than slinging code and too many managers more interested in expanding their own rice bowls and looking important than in actually streamlining how stuff makes it to prod. And then you need a pile of PMs or Scrum Masters to sort out the mess.
It may be obvious to conclude that good programmers are good reviewers but I think all variations exist. I wonder how many people don't like their jobs anymore because they don't like reviewing code written by AI.
Now I can work on much larger and much more challenging projects where I care more about the end result than learning little details.
Something has certainly been lost, but I’ve at least gained something else in return. I can also do use an empirical approach now, like testing whether a big change would be beneficial without having to map it out and try to estimate the effect. This has made what I try to work on more research-like than engineering-like; I’m trying to figure out if something is possible rather than how to best implement something I know is possible.
I still worry that soon there won’t be any way to contribute economically, that the machine will eventually be better at everything, but this is a much broader problem than just software or even engineering.
If anything, I enjoy programming more now with AI assistance, not vibe-coding. The depth and breadth of my work has increased greatly (yes, I still read and write a lot of code and documents, that the agents build on); I'm like a kid in a candy store.
Projects that would have seemed too daunting now seem feasible to tackle. I could even get started on implementing my dream language, of a knowledge management system for agents, or ... The possibilities are endless.
Now, my software engineering is for my own company, maybe that makes a difference. There is no one breathing down my neck to use more AI where it does not make sense, or to use AI this way or that. I am in the driver's seat, and I only delegate to AI what I need it for, and what it is good at. There is still much that is required of me that the AI cannot do, not anytime soon.
That is 100% the difference. Most people working as employees aren't passionate about the company as an entity. They won't love your baby like you do. What they did love was taking your baby for walks; now they're left with just changing dirty diapers.
For what it's worth, I also love coding and have been coding since I was 5, but the writing is on the wall. The stuff I get done in a day now would have taken man-years before. It's like having a genie at your disposal.
I have always been the opposite, despite mostly having IC roles in my career. Still, I don't enjoy the change, mostly because the UX of this tooling is absolutely miserable when you need to do anything serious with it. I am past the point where I thought it was user error and think it's just fundamentally baked into the ecosystem to be annoying to work with. At least in most management situations, the employee isn't going to gaslight you, question every tiny decision you make (or worse, subvert ones it finds more 'favorable' for itself) or just make up wild, easily disprovable bullshit in a constant stream of verbal fecal matter. I've worked with maybe one person that came remotely close to that and they were easily ignored. This stuff's persistent and unavoidable.
I am getting a lot more done, but I am a lot more tired.
By far my biggest use at the moment is just getting all the software I use to function as I want it too.
I'm much more on your side personally though, because first, when you know there's a better and faster way to do something it does take away from the enjoyment of doing it the old way. Some of the enjoyment for me came from being unusually good at something vs. everyone else.
And second, even my hobby projects are always something that I want to make real, more than they are something that I want to do just for the fun of working on them. Even though I enjoyed the programming and puzzle solving process, the end product is still the main goal. So I'll be using AI at home as well (and it is nice to reify projects faster than before).
Having said all that, I recently read Use Of Weapons (1990) by Iain M. Banks and this passage encouraged me about the future:
AI excelling beyond a human at writing code is not an objective truth.
I would say even without AI, a manager telling a developer what to code or how to code, will never be better than that engineer themselves being the driver and the developer.
For me personally, quality code has a strong human component. Only I know what's best for the code, based on:
1. my deep understanding of the long history of the organization and of its mission and concepts 2. my intricate personal relationships with the users and the other developers 3. all of my hundreds of unique, personal experiences troubleshooting problems, which are always a combination of code and interpersonal relationships, the open source community, and opposing stakeholders.
Until AI becomes AGI (and that's a big if) and you can train it on 43 years of experience, it will just be an approximation of a true person's holistic engineering pedigree. It will always be asymptotic to true competence. It will always be second-best, even with a human driver.
And the skills they provide are transferable. Patience, attention to detail, staying on task.
One step closer to the matrix. Send the white rabbit over, I've got your floppies ready to go. No ai, and developed without an active internet connection, so you can listen to music safe in the knowledge that your music isn't listening to you.
There's also a tension in your comment. Are we prioritizing "advancing society" or programmer feelings? If it's the former, then it doesn't matter if AI is better than humans, whatever moves us forward faster is better. If it's the latter, then we're talking about holding back progress for the sake of a programmer's feelings.
For the record, I'm a lifelong programmer who loves AI because now I can experiment with new ideas at a scale I could not before. I feel zero loss. In fact using AI brings back the magic of technology that we lost in the 90s with the "business-ifying" of tech.
University is not easy, so how did they skip the easy part? How did they get good enough grades everywhere else to pass?
Turns out: They must have done something that has nothing to do with programming but it is still computer science. What was it? What did they learn???
At the same time, I do wish it wouldn't do my proofs as well lmao
I don't agree. There are many creative and useful fields: engineering, law, finance. And others that don't require as much creativity but may be enjoyable due to social factors: medicine, teaching, emergency services.
The dilemma that people have faced is that they don't want to do earnest work, but still need money to live. There are enough types of work that are useful and stimulating in different ways that everyone should be able to find one, so long as they are given opportunities as children (we do a decent job of this in the developed world). At least, that was the contract. Many fields, not just software, are facing existential crises.
The second thing revolves around this:
> In fact using AI brings back the magic of technology that we lost in the 90s with the "business-ifying" of tech.
This is the complete opposite of my experience. AI seems like the logical conclusion of the business-ifying of tech. Get rid of all those pesky nerds, and the MBAs can just create whatever grift pops into their head.
I'd argue you're conflating two independent axis: technological progress and societal inequality.
AI is useful today and is getting more powerful and efficient. This is obviously great for humanity! It's enabling discoveries that were previously out of reach, and allowing us to accomplish the same amount of work with less effort.
AI is also generating massive amounts of money for the wealthy and powerful and increasing inequality. This is not great for humanity!
I do think it's important to keep in mind that this centralization of power and wealth is not caused by AI directly: the pattern exists whenever there's a desirable but scarce resource. The same argument that AI is increasing inequality can reasonably be made about concrete, steel, and steam engines. The problems caused by AI are really human problems at their core. It's possible AI could even help us fix them! (I'm not at all hopeful.)
>AI is useful today and is getting more powerful and efficient
No, it's getting less efficient at an exponential rate without proportional performance gains
>do think it's important to keep in mind that this centralization of power and wealth is not caused by AI directly: the pattern exists whenever there's a desirable but scarce resource. The same argument that AI is increasing inequality can reasonably be made about concrete, steel, and steam engines.
The goal of these LLM companies is to funnel all economic activity through their models. LLMs only exist because of how they strip mined human work without consent or compensation. I.e. LLMs do directly cause the centralization of wealth and power by how they were designed and how they are meant to be used.
While not an LLM, protein folding is now often done via transformer.
> No, it's getting less efficient at an exponential rate without proportional performance gains
AI is expanding in both directions. Sure, it's eating the world with its compute demand and the bitter lesson bites, but we're also seeing drastically improved efficiencies employed alongside. The cost per task is dropping fast.
> The goal of these LLM companies is to funnel all economic activity through their models.
That's what all companies want, AI or otherwise.
No arguments here on the ethics of AI companies though.
Take away intellectual labor and society becomes inherently less equal, i.e. feudalistic.
They have a hard time dealing with those people that tell them their ideas are bad, unethical, or illegal.
Now they have their Fisher-Price steering wheel hooked into the drive train of the business!
https://doctorow.medium.com/https-pluralistic-net-2025-08-18...
Programming is the art of automation, and is being taken to new levels. It'd be quite damming to the field if a substantial % of the world's output over many decade hadn't brought us here, tbh.
Were builders crying when the Hagia Sophia was built? Projects come and go, nothing is forever.
programming will likely be the same. fewer people who care to be craftsmen, more hobbyist, a few artisans who really dig deep and make new patterns (we still dont see AI sampling out of distribution).
>reduction of the job to a hobby takes that away from them.
this is your personal feeling. they never had anything in the first place, just that the market was valuing a certain skill for what it could output. the perspective you are posting is purely a construct.
I didn't mean to say sewing and knitting weren't impactful - on the contrary, they are analogous to programming: they were impactful but now they're basically hobbies. Maybe there are more developments left in the clothing making process - not sure.
Knitting and sewing were "mainstream impactful work" for a long time, much longer than programming was.
Even in something as heavily automated and mass produced as clothing, there's still a market for smaller scale, higher quality products.
These skills are all less valuable today, btw, along with roughly anything that would have allowed someone to thrive economically pre-2025 in spite of occasional accusations of high-functioning autism.
The issue is not that programming is getting automated. The issue is that the only skills that retain value are in low supply among programmers - and every other career you've considered retraining for.
Great time to be in sales.
I enjoy writing complex programs because I get appreciation for my work. I don't really care that much, but I'm sure we humans love being appreciated for what we do.
This AI shit has degraded most of what I enjoy. Artists, musicians, and writers are drowning in a sea of slop. Mathematics is looking like it won't have a human future. Programming is something you can't get paid for.
The future sucks.
I got into this industry because I enjoy programming, and I wanted to get paid for what I enjoy.
It sucks that the fun part of the job is gone.
I used to love taking the time to whiteboard my project, write the code in peace (with maybe a background show). Go through the steps of structuring, debugging, testing.
Now I can write a prompt to basically do what I do instantly, and now I'm the manager and tester of something I didn't truly even make.
So sure the upside is I can do more, and I basically have to use it because I can't even keep up with what people want me to do without it. And it also seems so silly to go back.
Everything fun is just gone.
Sorry but that doesn't apply to day to day work life...
That's the economy. It's not like all our problems are solved or everything that needs building is finished, even in very AI-forward tech companies. We are busier and working harder and longer hours than ever, and we've been hiring thousands of people in India.
I think just about everyone in the industry appreciated hand-typed programming as a job, because it was a pleasant enough craft to practice and for a solid couple of decades programmers were treated extremely well in comparison to the vast majority of other careers out there.
Today, juggling a bunch of agents to do what you want them to do, figuring out how to verify their work, remediating problems, and spreading yourself thin cognitively—all the while looking over your shoulder at an increasingly precarious job market—is notably less pleasant for a lot of people. It sounds and feels a lot more like an engineering or product management role—something a lot of programmers would never in a million years have signed up to do voluntarily.
I'm probably an 8 out of 10 on the cooked scale—I'm juggling a suite of ten iOS apps and haven't written or read a line of code in over a year—but I get where these people are coming from. I think the suggestion that they can reclaim what they've lost in this new job they've been loudly expressing distaste for by adopting a new unpaid hobby will ring hollow to them.
Look at Andrew Luck.
If people stop being interested in human-played chess, then yes he's out of luck. But that could happen with or without computers. Same for football and anything else.
So I don't think this analogy works either.
The important parallels of the analogy are that:
* Magnus dedicated his whole life to chess, just as many programmers have done with programming.
* Magnus enjoys the craft of chess, the human elements of creativity and cleverness, just as many artisan programmers do.
* Both prefer to be hands on and don’t enjoy being relegated to managerial / supervisory roles, and would probably move on to something else if they could no longer do the part they love.
* Many outsiders do not have any idea how much time and dedication it takes to reach the upper echelons of a skill like that, and struggle to empathize with those who have done so, and perhaps even feel Schadenfreude to see these folks “taken down a peg.”
Mathematicians have recently been lumped into this category of knowledge workers at risk. Many are grieving gratuitously and suffering a lot of flippancy from Schadenfreude-feeling onlookers.
A computer programmer, on the other hand, is paid to produce something. They aren't entertaining anyone (except themselves). The act of producing code doesn't matter to the person that wants a binary. The binary is what matters. So if a computer can out-program a human, then the human is in trouble. So it's different.
But, nothing is stopping a human from programming for entertainment purposes. And nothing is stopping humans from being entertained by watching other humans program, maybe competitively. Like chess, or basketball. Or from getting together and solving programming problems. Then it doesn't matter that a computer can run circles around them, and nobody can tell them to stop doing it.
I get it, though. Younger people in tech should be worried.
But I don't really understand the "grief" by people who say they're losing their hobbies. Like the top comment in this thread: https://news.ycombinator.com/item?id=50158159
If you don't understand grief then I'm wasting my time. Sounds like I got suckered into arguing with an AI yet again.
This is not what I said and you're being rude. I'm saying that I don't understand the grief being expressed for the loss of a hobby. Because as I explained, computers can't invalidate hobbies.
> Sounds like I got suckered into arguing with an AI yet again.
Nope. I'm a human and didn't use AI for anything I wrote. But thanks?
Making art is often considered entertainment, and it can also be commoditized. The difference between entertainment and work is often thin, dictated by personal enjoyment, not how one is paid. I can be paid for something I find very entertaining. I can also be paid for something I find dull.
If the thing I find entertaining disappears, it's reasonable for me to feel grief. For something I enjoyed is no longer available to me, at least not in full.
There's a community around programming, much like chess and basketball. Seeing that community deteriorate into solely programming for profit, and using AI to do so, is tragic. Programming has been very focused on profit for a long time, but now even those who think it fun are debating on whether or not to offload their work to a machine that does all the work for them.
Now all I have to do is make sure they're not peddling an actively bad idea and if not I can just tell the bot to incorporate their dubious ideas.
In the past it was all, I'm not familiar with this construct that solves our problem perfectly, why don't we cludge adapter pattern in here again.
This makes no sense. Everybody I know, including me, who hates the big picture of what AI has done loves programming immensely and does it as a hobby as well as a job. The people at my job who are most positive about AI are those who didn't particularly enjoy programming. You're applying Occam's Razor backwards.
However there are other folks in my company who, with very little programming knowledge, have produced viable, workable solutions for our company to use. All projects that were in my list of TODOs that I hadn’t gotten around to.
It was in that moment that I realized I’m very quickly going to be surpassed if I don’t adapt and move quicker with AI solutions. Since then my output has increased a bunch.
I’m not the most stoked on it, but as with all companies, results are far more important than the journey. So long as it works and is at least mostly maintainable, leadership does not really care how we got there.
But all of this was forcing my hand because people without the foundational skill set were creating equivalent output to my artisan, bespoke solutions. Writing is on the wall.
It's not necessarily a given that the person you're replying to agrees on the latter, and therefore they might not be stoked on you starting off by agreeing with them but going on to make an argument that supports the companies responsible for the very thing they're complaining about.
I think there's a contradiction in you stating that management thinks LLM-generated output is "maintainable". They went to business school, they don't know what that means. It's your job to explain to them the consequences of shipping code faster than the people whose job it is to understand it can do so.
The problem we’re collectively running into is that nobody knows if this actually poses a problem anymore. We’ll find out in 5-10 years, but businesses and clients can’t wait that long.
As a side note, I don’t think it’s fair to blame the “management” for setting expectations around AI use. They’re driven by market forces. If competitors are producing slop and customers are gobbling it up, then from a business perspective it _is_ a mistake to abstain from AI use.
What I’m finding quite fascinating is that we’re in a weird spot right now where we’re both expected to produce slop AND understand it / be responsible for it. Seems unreasonable.
1. The Decoherence Trap: https://maevyn.codes/decoherence 2. SlopCodeBench, a benchmark that quantifies how models decohere over repeated change requests rather than a single one-shot challenge: https://news.ycombinator.com/item?id=50108717
I think the most important takeaway is that it takes time for these problems to manifest. If you start with a particularly coherent codebase it will take years for slop code to degrade it to a point where engineers are noticeably less productive.
I didn't downvote you, but if I had to guess why you were downvoted it's because it's very obvious to many working software engineers that decoherence is a real problem. In many ways that's what this OP was all about; not just loss of joy in programming for its own sake but loss of the ability to design deterministic, fully-understood systems that won't require constant off-the-books maintenance eating into your free time.
Granted, I stick to frontier models and avoid cheaper ones, but if it quacks like a duck...
The last 20 years of experience have told me that that's the #1 variable that correlates with the death of a project, so I'm in no hurry to speed that along. People have been trying to find some way to skip that difficult step of actually understanding what's happened since programming was invented, and its always a bad idea - because that process of internalising a projects structure is the single most valuable thing for a project. Creating a project structure that can be internalised by human beings - and then doing that - is software engineering
But i get where you sre coming from
but guess what ???? most software don't
Timmy don't need to understand how landing page works in order to sell his cookie startup online
see where I am going ???? You are not get replaced
AI just replace people that work in "dirty and crude" work, think of its like wordpress plugin fixer from third world country
this is part of industry that you never touch and not want to work
Which is what we used to tell the junior to do.
My goal has always been to create something and use it. Ai still requires we do the intellectual side of programming, that interesting side of programming. I get to do more of that now that while the bot handles that annoying stuff. I'm a the happiest programmer I have ever been. I spend more time writing human these days, because the bots are bad at that.
I think I know people in both camps who feel both ways. I'd say it's all just noise at this point. What's clear is some people like it (for now) and some people don't (and probably won't ever). I've gone from liking it to not over the course of the last year. I used to love programming/the craft, and then got burnt out of it, so I loved it because it made building things easier again after I stopped caring about that, and then it made work hellish when everyone basically stopped caring about the craft at all.
But I also have found that where you work and what AI usage is like in your company is a pretty big deciding factor. Your comment, which I agree with, made me think about this. I consult a decent amount, and I've worked with companies where the engineering culture has adopted AI in a way I describe as "ugly." I should find a better word. But what I mean is that it's sort of like the attitude is "okay, now we can ship more tickets, so everyone take on 8x the workload". And they do a lot of "best practice" kind of stuff around using coding agents--every new anthropic article gets posted to slack, and there's always a new CI process for using agents--but they're fundamentally just trying to do the same work as before but kinda sorta faster. It's a weirdly miserable experience.
On the other hand, I've worked with companies where the engineering culture almost feels exuberant. I think a consistent difference is that in these companies, it feels like the engineers now have permission to try wildly ambitious things. It's not always optimally productive in the sense that more things get thrown away or end up just not really working out, but overall they move significantly faster and do more interesting work. Commonly, it's things they wanted to do before, but simply weren't realistic with their constraints around capacity. Or it's an idea like "this part of the codebase has always been a performance nightmare, but it's also a beast to untangle. I wonder if I could setup an autoresearch loop and let it run all day, even just to harvest some ideas?"
But yeah, when everyone sort of gives up and/or are forced to move an ungodly number of smaller tickets, it is a miserable time.
I declined manager position multiple times, now I am forced to it
And no, I can't code without agents as a hobby anymore. As a rational person, I feel like this is just useless nowdays. Huge part of the fulfilment that it was the most optimal way to solve the problem. Now I feel like I do useless work
At least two people I've worked with have become "AI Influencers" on social media. They were two of the least capable programmers I had ever worked with. Both super nice people, and I enjoyed their company, but both generated absolutely terrible code.
I think there's generally something to people who were sidelined by their lack of skill feeling empowered by AI.
I've never been the biggest hobbyist programmer. My hobby programming was a means to an end for the actual hobby project I wanted to use. I'm more of a "maker" than any one kind of engineer. That being said, it was still fun at times and those personal projects literally got me jobs and allowed me to bypass live programming tests because I was able to show I knew how to program as I explained every design decision.
LLMs sorta fucked me over. My strongest skills aren't raw programming knowledge or fact regurgitation. Personal projects were how I showed potential employers my actual end-to-end problem solving. The projects were never polished, they didn't have every error condition formally handled in a logging framework. But they organized ideas into data and made things happen.
All that is worthless now unless employers want to see my pre-war steel collection on old repos. I'm defeated. Using agents for personal projects feels like playing a game with all the cheat codes turned on. It stops being a game, it's pointless. Me telling the machine to make something doesn't prove anything.
I still want to remain housed into my senior years and this tech has possibly kicked that out from under me.
True, but dealing with agents is still easier than dealing with people (at least for introverted personalities). That being said, you still need to deal with other people's army of agents.
I see it similar to programming, before it was kind of a craft, many crafts are not paid very well, I know that woodworking can be very fulfilling but Ikea exists and noone pays you 1000€ for a simple table. But programming was a well paid craft because people are bad at thinking in general and not everyone can do the thinking good. Now we externalize the thinking and hope it will just work somehow... But the point is..that is a different job. Others already said it, suddenly you are a manager and not a craftsperson anymore. No more tinkering.
About the hobby stuff, I hate that the unannoited ones slop there way into _my_ hobby space, pollute everything with one shots that get abandoned in a few hours and drown everything with their half baked brain childs without polish that all look and feel so same-ish, the "map of something something" projects I have seen that were missing any kind of error validation or check of completness is astonishing.
As many people here and on Reddit can attest, I am one of these loud voices.
I actually see programming as all three: a hobby, fun and a profession. And AI itself is... neither. It's not enjoyable because you have to understand everything anyway if you want to do a decent job, no matter if for hobby or for work. The only thing AI can do is save you time when you have no starting point at all - i.e. reconnaissance in pentesting.
Now that you phrase it that way.. I think I'm very glad to have transitioned from 'programmer' or 'dev' to 'big-tech software engineer'. The jobs are very different. I had no idea what I was in for. SWE is much less fun. All meetings and alignments and docs. But I think that's what my job would have been forced to become with AI anyway, so at least it prepared me for this new world.
If that sounds like you, then don't be discouraged to keep thinking.
1725 points|df07|11 years ago|854 comments
https://news.ycombinator.com/item?id=11348798
Obviously entrusting an entity to do everything is different from using a dependency or library, but aren’t LLMs for work just a natural progression of this? Programming is about automation. Programmers have been automating and outsourcing their work since the concept of SDKs were invented. Or even since high-level languages were invented. The industry was already doing this, and doing it excessively, long before Claude. left-pad happened a decade ago and we’ve learned nothing.
>Second, even if the package’s logic is correct, I can’t help but be amazed by the fact that developers are taking on dependencies for single line functions that they should be able to write with their eyes closed. In my opinion, if you cannot write a left-pad, is-positive-integer, or isArray function in 5 minutes flat (including the time you spend Googling), then you don’t actually know how to code. Hell, any of these would make a great code screening interview question to determine whether or not a candidate can code.
> Finally, stringing APIs together and calling it programming doesn’t make it programming. It’s some crazy form of dependency hacking that involves the cloud, over-engineering things, and complexity far beyond what’s actually needed.
Obviously LLMs make the situation even more dire, but I think people are overlooking that the industry was automating itself out of programming as a craft even before them. It's just the automation involved reuse of the work of other humans.
This isn't just "a bit more automation". It's "now we don't need lift operators".
Of course, prompting an LLM to one-shot an app is a quantum leap in comparison, but c'mon, the whole direction of the field is towards automation, abstraction, convenience. It's always trended letting the programmer write less code. Have to deal with less menial tasks. So LLMs are just the latest manifestation of that motivation. And, perhaps in a decade or two, something will come along to replace LLMs, and this conversation will happen again, except even with more desperation and despair.
But while your example exists, I don't believe that's how the industry works, yet. They're not just allowing laymen to one-shot their production software. Certainly some teams may be trying that, and some indie devs have made good money already selling vibecoded software. But I don't think the technology is good enough to auto-generate professional code at scale, without human oversight. Give it another eighteen months.
Detractors say it causes needless suffering, accelerates global warming, bad for you in excess. Some say it simply doesn't taste nice, and are met with "Have you tried wagyu-7.1, it's much better than sirloin-flash-3".
Or if you prefer "have you tried the new iPhone, it's much better than the HP iPAQ".
Its something I've noticed in many folks. You'd say something like "well sure, did you try using a different model" and you will be met with screeching. Its almost as if they've shut off facts and made up their minds.
An example: a year back I wanted to show that AI doesn't hallucinate that much.
Here's like a 20 comment long conversation with a person who believed that ChatGPT hallucinates easily. It took them a long time to prove it - they used toy examples and it did't work because they'd never used the thinking version. Then they had to resort to esoteric examples in the benchmark which still didn't work.
https://news.ycombinator.com/item?id=47693707
We get it, you’re vegan, but your agenda is extremely off-topic in this case, to the point of being flamebait.
But I would prefer to have a job.
Normally, such new tools weren't a perfect fit for automating shilling and didn't involve trillions of dollars on the line.
So AI is kind of just bringing that to its logical conclusion.
Oh if we only had a middle ground between hitting the slop button versus hitting every key manually.
> Yes, sure, maintainability, yada yada. For most code out there, that never mattered anyway.
I’m so glad we have an expert who knows what’s best for “most code out there” gracing this thread with his presence.
Great to hear that the author found peace somehow, though I would advise not to worry too much about what you consume out there if it affects you to the point of needing justification.
A reply I saw bluntly stated their opinion which could be summed up as "Good, ignorance is bliss".
Whatever your thoufhts are on LLMs, it has never been more apparent that HN is not a place for hackers.
I'm sure some SV tech fag who was little more than a CRUD monkey a couple years ago will show up any moment now with their own definition of "hacker" as tends to be the case on GrifterNews.
Fully agree. It's still (or even more than ever) more entertaining than watching Netflix. Why should I care if other people use agents to do the programming?
I've resolved to just building stuff for my own businesses and myself for fun. If an opportunity or enlightened soul comes along that wants to benefit from my knowledge, great. But I think it's best to just let people feel the full brunt of what's coming. Sometimes the only way to learn the lesson is to stick the fork in the socket.
Experiencing this led me to feel burnout. It's horrible. I now value thinking and the ability to think much more. Not that I didn't before, but I mean I have a whole new appreciation for it and a whole new disdain for Claude slop where someone could have otherwise done a better job.
However, the real value is in complex, non-trivial, and above all novel code. This has been true for a long time, and it's just as true in open source as it is in industry or academia. Here, agents are far less useful. In fact, the only people I've seen have any kind of success using agents on these kinds of "serious" projects are people who are already serious programmers. They already know how to design and test stuff, they already know how to supervise junior developers, they already have an intuition for which aspects of the app the can safely delegate to a less experienced developer. So they're able to use agents as a "better autocomplete" in their own code, or use it to flesh out the non-essential and more routine aspects of the project, or even just delegate the agents to the role of code reviewer.
Here's the thing though. The less experienced ones, the people who don't already have those skills, they do get burned when they "stick the fork in the socket," as you put it, or in other words when they try to use agents to take on projects they don't fully understand... but they don't learn anything! The ones I've seen, anyway. Relying on agents insulates you from the code, which also prevents you from building a mental model of what's going on under the hood, which in turn makes it impossible to learn anything at all. They just keep yelling at the agents, telling them to fix stuff even though each fix introduces yet another regression in an endless loop... they don't know how to dig themselves out of the hole they've dug for themselves. And after all, what is the solution? To spend five years down in the code mines, paying your dues as a junior dev, reading and typing and thinking about mere code for the requisite 10,000 hours? Nobody has the time or inclination for that.
Agents, in other words, break the learning loop. You can't learn to program, and therefore can't learn to design or architect software, not really, if you're using agents. I am unfortunately coming to the sad conclusion that the last programmer has already been born; if you didn't have 5+ years of experience and a pretty good handle on the more advanced aspects of programming (such as software architecture, design patterns, optimization, debugging, etc.) before agentic programming became a thing, then I don't see how you ever will.
It'll be like COBOL programmers - the industry will shrink from the top-down, with the first to go being the new college graduates looking to enter the workforce. Inexperienced devs have always been given less interesting projects, but they stuck it out because they were learning valuable skills, building experience, making the economically rational decision to invest in their own career. But anyone entering the market today will never get out of that trench - anyone that actually needs a real programmer for a serious project will hire someone with decades of experience - there will be plenty of supply. And the projects on offer for junior devs won't teach them how to program, only to supervise agents. As a knock-on effect, smart young people already see the writing on the wall and are pivoting to other majors, which just reinforces the trend. This might cause problems in a few decades when all the real programmers start to retire, but when has the market ever considered long term consequences?
Phew!
I'd love to genuinely love these blog posts but they are all the same! They're all like "now that machine are better than us $xyzargument". I'm so pissed off when I read so blantant lies like AI "writing code better than me" maybe better than you! But in general it's just a big lie.
> the economics of purely hand-written code have changed
who are you to know? I still hand code and produce a lot of value because I do care and polish everything I do. There are some pieces of the software that I wrote that I never touched again, they work perfectly, no errors no customer complaints
On the other hand, if you mean tight, performant, simple, well-factored code, then yeah. LLMs simply aren't there, and may never be. On the other hand, maybe that matters less than we thought it did. Time will tell.
To me it's just crazy of how dumb it is to test an idea with a vibecoded shitty app. I'm not saying it should be perfect from day zero, let's pretend we are talking about cakes, how crazy it would sound you will tell people to try and judge your new idea of a cake you don't even know how it taste or what's inside?
It's disappointing that the prevailing sentiment now is that architecture is less important than we thought. Software is always going to be getting more important, and all of it should be well thought out. If these people had these tools in 2004 they might've made thousands of fragmented buggy social media startups and actually slowed progress on that front.
For the first group, it doesn’t matter how good your cake is. If you spend a year perfecting it, you won’t sell them any more cakes. And for many products, this group is basically 100% of customers.
The issue is nobody is spending time in review because the stuff coming out "just works". So if no one is reading the code, what is the big deal with naming? The big deal is we need to know what our later sessions are talking about and we need linguistic separation of concerns.
A higher level form of programming, if you will. Linguistic precision seems to be where the specificity can be lost or won. The naming will reflect that with some degradation. But humans will also degrade the spirit of the specifications as well in many cases!
Hopefully this post makes some sense to someone.
I mean, who's doing the factoring then?
Also "great point" is sadly a phrase that has been co opted by sycophantic AI so I can no longer take it seriously.
I see it as a multiplier (the bicycle metaphor) and yes certainly getting across town on a bicycle or across the country on a plane is less cumbersome on your health than walking every where.
I don’t ride motorcycles, I rode a scooter for a long time doing a delivery job as a younger man and the speed and instability (of two wheeled motorized vehicles) along with lack of visibility to others creates a dangerous combination.
And yes I can see the corollaries there to AI usage especially in these days when the excitement of the acceleration is being found seductive by many and without much visibility into the longer term consequences.
People used to be able to write code that was horrible by all the standard metrics but still "just worked" (or at least appeared to). From the stories I've heard, it would get into production all the time, too. But if it's a bigger problem now, then necessarily we know how to fix it, so.
I guess the excitement of the new toy is just leading to some irresponsibility.
But we don’t always do that and then things get rewritten and broken etc
Welcome to life in software dev 2026, not so different from 1996
There are a lot of mediocre programmers aren't there, I wouldn't be surprised if on average Claude is doing better than what exists. It's not a "lie", anyways, just a point of view.
I'm ok with vibecoding scripts, I love programming the machine and making the end user happy not dealing with bash! I don't judge scripts quality as long as they are tools in the workflow, I dont care about that, unless they will somehow be part of the production. The name Script implies that, there is architecture and Claude Opus sucks at architecture
I don't begrudge the business for making the prudent choice of making me use AI, I just sort of lament that what I spent years enjoying and getting good at was so utterly derivative that it was perfect for the LLMs. It would be insulting if it weren't just a fact.
Just like if you don’t read you will lose the ability to.
These skills are not innate to us and can be forgotten over time.
I wonder if it will get to the point where people don’t even know what to ask for. Or how to judge if what the machine produces could have been better. Or if it does what they think it does.
This is wholly untrue.
For me, it writes code that mostly works, which is great. It insists on adding some of the least readable comments I've seen. It also likes to pad out the code to about double or triple what I would write.
It doesn't seem to have that little voice that says "hey, this approach sucks" so it needs the harness or the prompter to tell it to go back and refactor. Even Fable and Opus struggle with this.
Well, yes. For humans, the "this approach sucks" voice is still very separable from the thought process that just figures out the syntax and algorithms etc. The important thing is whether the overall system has this feedback and that it's having the desired effect. Better harnesses are progress in the same way that more capable models are.
And yeah, there's a wide range in the standards that programmers hold, in practice. Plus they don't at all agree on what's "better" than which.
The general human approach to programming at most companies I worked at prior to 2024 was to head off in vaguely the right direction and refine the approach as we went. There might be a design and a plan, but at some point we just winged it and fixed our work after the fact.
This seems to be the worst way of getting an AI to do anything. They do better if they can one-shot an implementation from a clear spec then only need to fix bugs. Better models can handle a less clear spec but they still can't really iterate in a major way. To have an AI model build a bigger thing, you one-shot increasingly detailed specs and decompose them into smaller ones.
And even if they somehow got better taste, it wouldn't be your taste. If that were acceptable, why bother prompting anything in the first place? At that point the system might as well choose its own projects entirely. But I don't think humans would want to live in that world; even if not paperclip-maximized, we'd eventually become an underclass.
That's assuming basically unlimited capacity for the models to get more powerful. If there are limits (and I hope there are!) then it'd be even more certain that projects continue to require human guidance.
But even despite that, I hope I don't sound AI-psychotic if I suggest: the models are capable of at least assisting with that spec-writing and decomposition.
I completely agree. The fact that the software from places that are very vocally pro-AI (microsoft, salesforce, etc...) is still so full of bugs is really contrary evidence.
The evidence that the AI tool is better than oneself? I don't know about you, but I think I'm a pretty good judge of whether Claude Code is better than I am, having seen my own work and Claude's firsthand.
Those weren't lies. The AI does produce better code than those people. They simply weren't very good programmers to begin with.
I think this goes a long way to explain why there is such a huge difference in opinion about the quality of AI-written code: Some people say LLMs write code far better than they could, and others say the code is always terrible and requires a great deal of fixing before it's acceptable. How can two different developers say those very different things about the same product?
It might be a tough pill to swallow, but if you think your LLM is producing better code than you could produce... well, I might have some bad news for you about your own skills.
But in the new environment of LLM-assisted coding, the LLM produces acceptable code so much faster than I can write it. Just taking time to understand and retool the LLM's output means an order of magnitude lower productivity and it doesn't feel like that's acceptable in this era.
Folks I know are landing 10+ PRs a day and nobody is reviewing anymore, and these are not companies that would have been skipping code review 2 years ago. The whole culture is turning against taking the time to comprehend and reason about our own work.
There is a wide range of code that will do the same job. 15s vs 2h for code 60% as good? 10 minutes for code 80% as good? 10 minutes today and 10 minutes tomorrow to fix something that is 30% as good?
A lot of people are going to draw that line in a different place and now that people can make that trade off you will quickly find they will and it probably wont fall where you would like it to.
At the end of the day a business pays you to produce value that will earn them money / customers. They don't pay you to write good code. Sometimes those two align and both sides are happy but they don't and often wont.
On the other hand, I'd also expect the most degenerate among us to not care about the uncanny details. So who knows, really.
I would say he is a good PM/visionary.
It's not a "skill or prompting issue", it's either you have a software factory or you don't and suffer the poor quality.
And that's just sorting out my self-image. Convincing someone who controls the pursestrings that I'm adding value was hard already.
Well the Dunning-Kruger effect could certainly explain the cohort that thinks they're skills are amazing and unbeatable.
"The AI labs are scared of their own models, surely they must be credible sources on how powerful those models are"
"They are asking the government for regulation, they must be even more responsible than the average company"
"They are claiming to be a software engineer who automated their job with AI, surely I have to buy their course"
There are many boilerplate pieces of software that an LLM can generate faster than a human, but nearly half a decade into the LLM era there is no evidence that it's leading to the apocalyptic end of software organizations. And that's all with the subsidy in place that is now disappearing. Everyone online is lying for money, and that money is coming out of the same investor-subsidized marketing budgets.
I would recommend "The Leprechauns of Software Engineering" by Laurent Bossavit for an insightful and entertaining overview of just how much of software "engineering" is based on "truths" that we don't question.
You’d have to be pretty unbelievably stellar to produce code 200x better than Opus. I can get a lot of business value out of not-great $20/mo coding. The people writing our checks do not care about the same things we care about.
And I say this as someone who is very upset about what this stuff is doing to my career and hobby.
But I read the OP's argument to mean that your responsibility as an engineer is now to understand the tasks an agent can accelerate. And in many cases that might be the task that you previously enjoyed, like working out some tricky variable length encoding serialization scheme, or tracking down some nefarious race condition by stepping through code and reasoning. (The eureka moment is a big dopamine hit for me, dunno about you.)
LLMs, even in a good harness, are deceptively simple, but in reality rather complex tools. They're good at grasping intent, but they can't read your mind, they might not have access to the full context of your business, and they require learning how to use them properly.
In other words, hand holding.
Your error lies in thinking Opus 5.5 is equivalent to a "minion". The "5 software developers" in my example were meant to stand in for "reasonably competent, above-average developers", and not that the architect is the only one on the team who can "think".
Of course, it may well be the case that you are working in an elite sector of the industry where even the bottom rungs are filled by the top 20% of software developers. Possible, but unlikely, so I'm still leaning towards "you're holding it wrong".
Opus has to be 200x worse for them to sell that subscription. If their AI is as good as they claim, they should charge way more, like one dollar less per month than a human developer costs.
Only if they had a monopoly, which they thankfully don't.
Yeah and this is what pisses me off about the software factory deniers. Without the software factory to guide the agent and reject its code, how exactly is it supposed to do better than an experienced developer who already internalized the company workflows? All the Agent sees is its context window, it's not a fully formed human that can learn the workflow over time.
If I guide the Agent file by file and function by function, then review and reject the code, the code quality becomes reasonably good, but when I let it do whatever it wants, the code quality reaches all time lows and it just compounds on itself.
I also noticed that the code quality heavily depends on the underlying software ecosystem with Javascript often having the worst code quality because it presumably got trained on minified code as well.
Agent harnesses do not solve software development in a hands off way. You still have to baby sit them. Every single time I use an auto review based workflow (aka vibe coding) it just doesn't work.
Then AI came a few years ago and that was interesting as a tool to play around with and do some simple small code (snippets) generation or as a buddy to discuss problems and find solutions. Meanwhile I was getting more and more bored with implementing stuff that I did already so many times at previous jobs. Same principles, different code base.
Last year I started at a new job and was first hesitant, but I decided to go 100% AI with doing a rewrite of an existing application that will be phased out. So actually I could develop from scratch with the help of AI. First I copy/pasted code, then this year I found out about AI agents.
Never in my life have I been so productive in creating features/functionality while the code that AI generates doesn't differ that much from what I would have written myself. But then in a few prompts and 10 times as fast. Including unit tests. Which at many places that I worked were more often not a thing, then that they were.
Looking back at my 'career' it also made me realize that I do like "tech", but don't like coding as much as other software developers, I guess. I really like AI to make functionality and things working in a very short timespan. I enjoy 'AI-coding' much than I'd have thought. So many things that you would not even try because it would cost too much effort, time or money. Now I can test within a morning a few different features to see which one fits best and works best.
Never have I been such a happy 'developer', I don't miss the coding part. And AI will only improve, so personally I don't see a bright future for a lot of software developers to be coding in the traditional sense.
How do you know that? Do you read AI code? I heard (DHH, Uncle Bob, my own co-workers) that reading the code is no longer an economically viable activity. And if you are not reading the code, how would you know what it looks like.
Yes of course. But sometimes, when no one is watching, you can quickly sneak a peek at the code changes. It's ok!
I claim that AI "writes 80% of my code". But realistically I am in the hot path of my agent ("cd repo", "claude", "do X", "fix Y", "no not like that"), and that's essentially equivalent to writing by hand. I am afraid others will soon see through my charade.
That's the reason why I still try to inspect the generated code to see how my problem was solved and if it reads plausible. There are also moments that I am surprised by the implemented solutions which look very elegant and "smarter" than what I could have come up. These are also the moments I learn from AI. I can even ask AI why it did things certain ways and it gives a nice summary of arguments. But I am often also critical towards AI if I am in doubt, resulting sometimes in a sort of discussion and sometimes I can convince AI (with more context) that the solution it created should take another route. It really feels like working with a really fast co-worker. And on top of that, I am better in writing up my thoughts to discuss things instead of talking with someone. Doesn't mean I don't discuss in real life with co-workers, but it is so easy to use AI for it too.