I insist on reviewing every line of code. But in the last few weeks I realized that LLMs are simply better than me at reviewing code too and I'm just finding excuses to not do the real work, which is to write more prompts.
I find LLMs are better than humans at finding bugs in code review. But humans are bad at finding bugs in code review in general. That’s not what a good human code review accomplishes.
A good human code review takes into account the broader business context and the structure of the codebase. It finds instances of poor design or readability. It suggests patterns or existing utilities that the author didn’t make use of, and points out antipatterns.
LLM reviews are a great first line of defense against certain classes of bugs. But I consistently find major issues in code review that the review bots miss — and some that they couldn’t have caught under any circumstances.
Just last week: 800 lines of unnecessary (and clearly LLM-written) code with high cyclomatic complexity that could be easily replaced by 50 lines of mostly declarative code by using existing patterns and utilities; several missed opportunities for converting incoming data to native types up front upon receipt and enforce stricter typing, rather than passing around things like dates or UUIDs (or even basic numeric types) as strings and trying to parse them over and over; and a case where the author had misunderstood which of two possible sources certain data should be derived from, resulting in subtly incorrect data being saved to the system of record.
A phrase I've been toying with is eyes free vs hands free coding.
The distinction is important. While I too hardly write code these days, I do continue to make smaller changes and look at what's being done. This is as opposed to the group who has decided vibing mega commits is a good idea.
I am not doing small changes though, I add large features or refactors daily, which involve changing hundreds of files and thousands of lines, even if I wanted I couldn't review everything by hand.
So you're saying simultaneously, the code is "higher quality than ever", but you also haven't even looked at the code to see its quality. Classic. I think you're mistaking quantity for quality.
If a product functions correct and performance is good, and AI agents write and maintain all the code, no-one has to care about human level code quality. The concept of "code quality" was only invented to make it easier for humans to understand and change code.
That's a really good point. Most good or bad code, and all the best practices came usually from the perspective of having the code be readable and maintainable by humans long-term.
And performance is easy to test automatically by having performance benchmarks automatically ran on deployment.
Even if that's true, it doesn't follow that you personally will make the world any better by refusing to use LLMs, given that you'll use them carefully and conservatively. And as for what other people do, you can't really control that.
Yeah, that's definitely true. The world is now moving so fast, and everyone feels pressured to match the speed of others, to not be left behind.
I do think/hope that as LLMs improve, that improved quality would come by default, or that people would still compete in quality instead of quantity, but this is also the consumer's job to not accept shit quality products.
We use AI to write almost all our code in a fintech - trading applications. If the quality is low, bugs cause monetary loss. Degradation in performance causes monetary loss. And so on. It doesn't happen. We have very good functional and non-functional testing and the coding agents are able to iteratively add functionally correct features and improve performance.
I don't understand what other people are doing differently to get worse results. We cannot ever go back to humans writing this code, because our competitors will eat our lunch if we do.
> I haven't touched a line of code in over 6 months, didn't even look at it, but the work I do is faster and higher quality than ever.
Well that doesn't make sense if you are writing code that is going to be used on critical infrastructure.
Imagine a disaster were to happen and you said to the auditors that "I don't know what I am doing nor did I look at the code" and you cost the company tens of millions of dollars.
Sure if your building toys, that's fine, otherwise, why take the risk in being liable for not looking at what the code does and then believing it is high quality?
Indeed, I am just building apps people use, they are not "critical infrastructure" and the world won't end if there's a small bug, I don't vibe code self driving cars, defense systems or medical devices. A
If there is a bug, it will be fixed quickly anyway, so only a few users might ever see it, or see it only for a brief time.
Even when manually coding, projects still have issues and bugs, I don't think much changes in that aspect now, if anything, there are fewer bugs now in most high quality apps than before, due to AI catching more issues. See all the vulnerabilities found in old core code and libraries nowadays, that humans didn't find.
LLMs write better code than humans with fewer bugs. They also do it much much faster so you can spend all the remaining time testing. Tests are written faster too so not only you have more time for testing but also you can use it better.
>>Sure if your building toys, that's fine, otherwise, why take the risk in being liable for not looking at what the code does and then believing it is high quality?
>>Don't you see the contraction you just did there?
I wouldn't trust anyone who refuse to use LLMs for programming. Their code is worse, they test it less and it takes longer to write. For me "no LLMs" attitude is a sign of incompetence. Good for hobbyist trying to keep their brain active but that's about it.
That is even more so for critical infrastructure. Just look at the mess human coding is when it comes to security. LLMs are fixing it at the pace we couldn't even imagine a year or so ago.
> LLMs write better code than humans with fewer bugs. They also do it much much faster so you can spend all the remaining time testing.
So does that mean you should never look at its code?
If a fault in the software occurs even after testing and there was an investigation, will you admit in the postmortem report that you did not look at the code at all and blame the agent for the error?
> I wouldn't trust anyone who refuse to use LLMs for programming.
Did I say I refuse to use LLMs for programming? I am telling you to check the agent's work.
I would not trust anyone who DOES NOT check the changes of an agent's code when using them for programming in highly regulated industries or in critical systems. For most companies in the Fortune 500, it is irresponsible to not check.
> That is even more so for critical infrastructure. Just look at the mess human coding is when it comes to security. LLMs are fixing it at the pace we couldn't even imagine a year or so ago.
No-one argued against that and it does not mean that you do not check with the agent at all even if the agent believes that they "fixed it".
AI is a hammer, problems are the nails. At the end of the day, you still need someone to swing the hammer. Aside from the occasional overstretch of scope and “reining it in”, the real bottleneck is humans.
We have more freedom than ever to build whatever we dream of, you still need a dreamer. And that’s why the profession of coding isn’t going away, you still need people to turn ideas into reality. I find AI soulless, it’s great at mimicking but is incapable of coming up with an original idea that wasn’t yours regurgitated.
When am I (as a user of a lot of software) going to start seeing any results of this unprecedented productivity gain?
This thread is full of 100x supercharged programmers, so why is it that I can't name a single useful feature added in the last 5 years in e.g. macOS, Windows, Firefox, Chrome, YouTube, VS Code, YouTube, Slack, Photoshop, Word or any of the other 5 programs currently on my taskbar?
The productivity gain is invisible using your metric.
What I have seen is that, with LLMs, an entire class of process brokenness disappeared in the in-house codebase. Before, there were many trivial tickets that, in the end, required a 5-line bugfix sitting in the queue for months. Now, such tickets are resolved instantly, and the backlog is shrinking.
Because the code is and almost never was the problem. Product design and business involves a lot more than code.
Personally, I did see a lot of entry-level products having a lot higher quality and better UX. At the same time, most companies go for fast iterations and are profit-focused, do product quality is not really a target for many.
The productivity gains will go where they always go when a mature industry undergoes automation. We'll reduce headcount, ramp up the output expectations of anyone lucky enough to keep their job, and expand the operating margins for shareholders.
None of this capital has ever been earmarked to build better consumer utilities.
- Try to prove the concept early on. When you start spinning in circles, then it might be time for a different approach
- I’ve had a lot of success with projects that have not started out as LLM codebases. And adding features one by one and testing really carefully by hand
- UI generated by LLMs are overly verbose. The key lies in the iteration. What you can cut, move or approach differently. But you have to have taste.
I have a hard time understanding that position when it comes to programming. What is "human" about typing instructions for the computer into a terminal? Was perforating paper more human?
When it comes to generative art, yes, that's a much more significant issue.
Not OP, but all the choices made along the way are the human part. The understanding and creativity to solve the complex challenges that arise in an efficient way that fits into the structure of the greater program.
Programming never has had one right way to accomplish a task, and that is disappearing with agents. Why would pick Ruby for a project because it’s nice for the programmer, when the agent is the one coding? In fact, Why would anyone ask an LLM to write a new program in almost anything other than JavaScript and Rust?
I'd say they took the inhumane parts of being a programmer away and people who spend 30 years turning themselves into robots don't know what tondo with themselves now.
Programmers were just the first shoe to drop. The same is now coming for the entry-level engineers as agents learn cad programming which has the same complexity barrier to entry.
We will soon trade that complexity barrier over to the agents and we won’t know how it is done, because the agents always did it for us.
I feel like LLMs are akin to a performance enhancing drug that's gained mainstream acceptance.
Everyone can run a 5 minute mile now without training. I used to like training to be able to do that, but now that it's accepted everyone can do it for free (and people don't care if you did it the hard way) the joy has been sucked out of it a bit.
It's more like they are being carried in the arms of a big robot who is doing the running for them, but everyone has somehow agreed to say that the human "ran" a five-minute mile because they ordered the robot where to run.
Except performance enhancing drugs are banned from sports because that’s how they want it to be played (it has nothing to do with safety, look at non drugs like the LZR Racer swim suit).
If you wish to make software engineering your job then it stops being a game where we ban things simply because they provide an “unfair” advantage.
If you wish to play at programming “the way it used to be done” then by all means go ahead. But mixing methapors between sports and work is not helpful.
I guess I don’t understand the motivation you have for perusing your hobbies.
To me, the effort part only feels pointless if you want to accept LLM-derived code as comparable. Any marathon run is pointless, you go in a loop and we have cars. The point is to do the run yourself unaided.
That a computer can play a virtual instrument better than me (and continuously better than anyone) has little bearing on whether or not I continue practicing on my piano.
I think the analogies are not helping us. This started by comparing LLMs to performance enhancing drug being used within the same sport.
But software engineer, as a job rather than a hobby, has always been about results. Hence my original point.
Hence my confusion. I am not perusing my hobbies for “the world at large” I’m perusing them for myself; being able to drive 26 miles makes no difference to me nor to the others who wish to run it.
I think many people derive joy from hobbies because they get a unique skill as a result of time invested. If the Philadelphia marathon started giving out medals to people who drive the course in a car, I think less people would want to run it.
I agree, but they would move on to another marathon that didn’t hand out prizes to drivers.
People who continue to do the hobby without LLMs will still get unique skills, like woodworkers who don’t use power tools.
I understand that this is a loss for those who were pursuing programming as a hobby only as long as it gave them a unique skill and made them feel superior to others. However, for better or worse, I can’t bring myself to care about that.
Hobby programming without LLMs and you gain unique skills (you can do the work without power tools).
Hobby program with LLMs, you gain unique skills again (you learn how to properly use these models and get to quickly iterate on your design skills). Regardless of what the big labs are saying, there is still a lot of skill involved in creating with them; if only because they don’t read minds so you have the get good at describing your taste and requirements.
I guess I fundamentally disagree with that being a good metric by which to pick hobbies. I am not picking mine based on some economic expected value calculation, it's whether or not I enjoy doing the activity for its own sake.
As an illustrator/animator who codes as a hobby and to produce games that are mostly just making that art interactive. I feel like AI is taking over my humanity when it’s doing any visual art or design work. But when it does code. I care much less and am just happy that I reach my goals faster.
This is very personal and I completely understand that if coding is your main thing, code by genAI feels worse.
I do fear for the future…
At least coders have been able to save some money. Visual artists are mostly broke.
> I feel that LLMs and their overtly verbose output trick less knowledgeable programmers of the quality of their output. When Claude spits out refactored code with tons of comments, documenting each method and variable, and tests for everything as well, it looks impressive, at a first glance. But if you dig deeper, it's sometimes just complete bullshit.
This is pretty spot on. But it's important to note that the resulting code, is still often significantly better than what those developers used to produce. Especially under tight deadlines and pressure to complete CRUD like interfaces. We simply can't afford spending a whole day tweaking a sortable table on the web anymore.
I agree that it produces code that is better than many of my colleagues. I have spent the last 14 years honing my ability to build code in a scalable way that works for basically any domain. I still produce better output than the LLM. But the incentive to encourage my colleagues to reach that kind of level is just gone, and that's what frustrates me. As humans, we could have been so much more efficient, but instead LLMs came in and are steering us towards the median of slop code. Slop by my standards anyways.
So I have some questions for agentic dev aficionados:
- Why should I even start vibing up my desired tools and projects when I can just wait it out so a better clanker emerges 6 months down the line?
- What guarantee is there that Rust is the best programming language for vibe coding when the models could improve to directly output assembly? What incentive do you have in vibing up something?
I can answer question #1 from the perspective of an employee who trades time for money: I don’t get immense joy [1] from writing code to do whatever dumb corporate thing my employer needs this week.
An agentic workflow lets me literally sit on my hands (or play Star Wars: Galactic Racer, or play my musical instrument, or work on my CAD hobby, etc) while the agent machine does my work for me.
The best part is that this is how my employer wants me to work. They literally want me to be lazy. Works for me!
Where is my incentive structure to not use AI?
[1] coding manually brings me a non-zero amount of joy and mental satisfaction and stimulation but I am still trading time for money.
Yes, but if I only use the excavator for 5 minutes and coast the remaining 55 minutes, I'm less useful than someone who uses the excavator the whole hour. Now, if I'm particularly skilled at using the excavator compared to my peers, I might get away with working less. If not, working 5 minutes while charging for 60 minutes will eventually get me fired.
Generally, automation does not make us work less because the expectations rise equivalently.
I think you completely missed the point. If the work becomes faster, you need to get more work done per hour than before. The work has been redefined, but so have the expectations.
1. At my company I am basically a regulatory requirement.
2. I’m paid to be a warm body who is available. That’s I am paid salaried instead of hourly.
3. I don’t really care if this situation changes in the future. If AI takes my job there’s nothing I can personally do to stop that from happening. I’m just saving my money and planning to retire as early as possible. I view work as a necessary evil.
I'm not writing a blog post just for this, but the man is right. For me, the only thing I expected from a LLM is a StackOverflow/Google that actually works and maybe some static code analysis.
> models can write code better than the average engineer
If you're not interested in "average", volume isn't going to make up for quality. I'm certainly not perfectly satisfied with LLM output quality, but even the best programmer in the world will undeniably be helped by LLM tooling, even if only for having it write tests, benchmarks, run experiments, create research summaries, and so on.
I guess nobody cares about, if some one use LLM or hand written code. They all care if you are hired and deliver the output comparable to others with all their requirement fulfilled. I still remember how wordpress developers were against elementor during its early days and look how pagebuilders became default choice now for wordpress.
You don't need to use LLMs for coding to stay "competitive" with your colleagues.
I work for a FAANG, we have unlimited AI credits to be used on any model we want, with basically any tooling we want.
My whole team massively uses AI, I barely do. I use not even 10% of the tokens that they do. Yet I have not only the highest rate of PRs raised, but also the highest rate of PRs I reviewed.
AI is all about fake productivity. You feel productive while you're achieving the same, if not less.
This is extremely accurate, but HN doesn't like it when you burst their bubble.
A good software engineer has always been able to write a ton of code all by themselves in a short period of time without any oversight. There's nothing romantic or intelligent about that. It's just reckless.
Teamwork is more important than ever, and yes of course that slows things down. That's the point. The decision making is the slowest part. 5 hours of meetings today can completely change the trajectory of the next 5 years. This isn't a revelation for anyone but the most naive.
This is a roundabout way of saying you're leaving the profession.
We now have people commonly running multiple concurrent programming tasks, while automatically and interactively testing them, reviewing them, and validating them. They're then handed for human review, where little is found because the quality regime is now set so high. We're talking weeks of development + testing/validating time brought down into the realm of days.
So this is pure "my horse & cart are just as good as the motorcar"-levels of argument. If I bury my head in the sand long enough, and deep enough, this will all just blow over, and we'll go back to doing stuff significantly slower/worse any day now...
The quality of stuff I am shipping today is the highest it has ever been, and tested/validated significantly more comprehensively. Why? Because time/resources was the limitation, not willingness or skill. I always wanted to test for every single possible edge case/scenario, but it wasn't in the time-budget (and diminish returns). Now I don't need to make that trade-off.
I've struck the balance of using it at work but not at home.
At work I still read through every line in the diff (unless it's a personal util/non-production facing/critical etc). Even with this I feel less 'in touch' with my work code than I do with my home code.
The comparison with heroin frankly is bizarre, as if using LLMs was purely hedonistic activity.
LLMs are tools like motor vehicles. They allow you to get somewhere, quickly but brutishly. You won't be able to appreciate the beautiful scenery much, as it rushes by. You're going to have to compensate for the exercise you would have gotten, had you walked, or it's going to be unhealthy. It's undeniably bad for the environment (electric or not) and has other negative externalities. The politics of many of the countries supplying most motor vehicle fuel to the world are beyond fascist by modern leftist standards. Yet, we depend on them, we have constructed a lot of the infrastructure we need to live around their existence, and even if every city became walkable over night, the last mile for most transport would be covered by some motor vehicle. There are many "what ifs" surrounding motor vehicles where the answer boils down to: "Well, then we're fucked". But we're not fucked today, and so on it goes.
Of course you can try not to partake in motor vehicles, and insofar as you don't make a moral grandstand about it, that's totally fine with me. But the rest of the world is in it for the ride. The first cars weren't very good either.
I've been suggesting my junior colleagues recently the following mindset:
for the last 5 years you have been milking cows, that was jour job, and you were great at it.
But this job ceased to exist [somewhere around August this year when Opus 5 appeared]
You have a new job: you are a barista now [and you are going to serve oat milk]
You can still manually type code, it can be a great hobby, but it is unlikely that someone will pay you a full time salary for doing it
Genuine question: Will you also reject the results and fixes regarding the current PHP CVEs until there are handcrafted versions? And - a question from another perspective - would you be open to using Mistral in a self hosted environment?
140 comments
[ 0.31 ms ] story [ 50.8 ms ] threadCool story bro, but it sounds like you are trying to convince yourself.
See you back here in a month, or so.
I haven't touched a line of code in over 6 months, didn't even look at it, but the work I do is faster and higher quality than ever.
Not sure how I feel about it
A good human code review takes into account the broader business context and the structure of the codebase. It finds instances of poor design or readability. It suggests patterns or existing utilities that the author didn’t make use of, and points out antipatterns.
LLM reviews are a great first line of defense against certain classes of bugs. But I consistently find major issues in code review that the review bots miss — and some that they couldn’t have caught under any circumstances.
Just last week: 800 lines of unnecessary (and clearly LLM-written) code with high cyclomatic complexity that could be easily replaced by 50 lines of mostly declarative code by using existing patterns and utilities; several missed opportunities for converting incoming data to native types up front upon receipt and enforce stricter typing, rather than passing around things like dates or UUIDs (or even basic numeric types) as strings and trying to parse them over and over; and a case where the author had misunderstood which of two possible sources certain data should be derived from, resulting in subtly incorrect data being saved to the system of record.
A phrase I've been toying with is eyes free vs hands free coding.
The distinction is important. While I too hardly write code these days, I do continue to make smaller changes and look at what's being done. This is as opposed to the group who has decided vibing mega commits is a good idea.
High quality code doesn't always lead to a high quality product.
And bad code doesn't always lead to a bad product.
And performance is easy to test automatically by having performance benchmarks automatically ran on deployment.
It seems most people are bad at evaluating the quality of their work, especially if they purchased it from Anthropic.
It could have also been a "use AI to test our product more", but I don't see that anywhere.
I hope you agree LLM's can actually help with quality, if used for that purpose?
And it's not just restricted to software. Everything I care about is getting worse.
Dribbling about how it could theoretically be better doesn't help undo the damage.
I do think/hope that as LLMs improve, that improved quality would come by default, or that people would still compete in quality instead of quantity, but this is also the consumer's job to not accept shit quality products.
I don't understand what other people are doing differently to get worse results. We cannot ever go back to humans writing this code, because our competitors will eat our lunch if we do.
Well that doesn't make sense if you are writing code that is going to be used on critical infrastructure.
Imagine a disaster were to happen and you said to the auditors that "I don't know what I am doing nor did I look at the code" and you cost the company tens of millions of dollars.
Sure if your building toys, that's fine, otherwise, why take the risk in being liable for not looking at what the code does and then believing it is high quality?
Don't you see the contraction you just did there?
Wonder if he tells his boss to "I pushed critical code to production without even taking a look at it" or is it just for show on HN
If there is a bug, it will be fixed quickly anyway, so only a few users might ever see it, or see it only for a brief time.
Even when manually coding, projects still have issues and bugs, I don't think much changes in that aspect now, if anything, there are fewer bugs now in most high quality apps than before, due to AI catching more issues. See all the vulnerabilities found in old core code and libraries nowadays, that humans didn't find.
>>Sure if your building toys, that's fine, otherwise, why take the risk in being liable for not looking at what the code does and then believing it is high quality?
>>Don't you see the contraction you just did there?
I wouldn't trust anyone who refuse to use LLMs for programming. Their code is worse, they test it less and it takes longer to write. For me "no LLMs" attitude is a sign of incompetence. Good for hobbyist trying to keep their brain active but that's about it.
That is even more so for critical infrastructure. Just look at the mess human coding is when it comes to security. LLMs are fixing it at the pace we couldn't even imagine a year or so ago.
So does that mean you should never look at its code?
If a fault in the software occurs even after testing and there was an investigation, will you admit in the postmortem report that you did not look at the code at all and blame the agent for the error?
> I wouldn't trust anyone who refuse to use LLMs for programming.
Did I say I refuse to use LLMs for programming? I am telling you to check the agent's work.
I would not trust anyone who DOES NOT check the changes of an agent's code when using them for programming in highly regulated industries or in critical systems. For most companies in the Fortune 500, it is irresponsible to not check.
> That is even more so for critical infrastructure. Just look at the mess human coding is when it comes to security. LLMs are fixing it at the pace we couldn't even imagine a year or so ago.
No-one argued against that and it does not mean that you do not check with the agent at all even if the agent believes that they "fixed it".
We have more freedom than ever to build whatever we dream of, you still need a dreamer. And that’s why the profession of coding isn’t going away, you still need people to turn ideas into reality. I find AI soulless, it’s great at mimicking but is incapable of coming up with an original idea that wasn’t yours regurgitated.
This thread is full of 100x supercharged programmers, so why is it that I can't name a single useful feature added in the last 5 years in e.g. macOS, Windows, Firefox, Chrome, YouTube, VS Code, YouTube, Slack, Photoshop, Word or any of the other 5 programs currently on my taskbar?
What I have seen is that, with LLMs, an entire class of process brokenness disappeared in the in-house codebase. Before, there were many trivial tickets that, in the end, required a 5-line bugfix sitting in the queue for months. Now, such tickets are resolved instantly, and the backlog is shrinking.
Personally, I did see a lot of entry-level products having a lot higher quality and better UX. At the same time, most companies go for fast iterations and are profit-focused, do product quality is not really a target for many.
None of this capital has ever been earmarked to build better consumer utilities.
- Try to prove the concept early on. When you start spinning in circles, then it might be time for a different approach
- I’ve had a lot of success with projects that have not started out as LLM codebases. And adding features one by one and testing really carefully by hand
- UI generated by LLMs are overly verbose. The key lies in the iteration. What you can cut, move or approach differently. But you have to have taste.
When it comes to generative art, yes, that's a much more significant issue.
Programming never has had one right way to accomplish a task, and that is disappearing with agents. Why would pick Ruby for a project because it’s nice for the programmer, when the agent is the one coding? In fact, Why would anyone ask an LLM to write a new program in almost anything other than JavaScript and Rust?
This is not an issue.
Nobody cares about typing instructions. If someone can click instructions or voice them, it all the same.
Programming is about finding the right instructions. Not about typing.
We will soon trade that complexity barrier over to the agents and we won’t know how it is done, because the agents always did it for us.
Everyone can run a 5 minute mile now without training. I used to like training to be able to do that, but now that it's accepted everyone can do it for free (and people don't care if you did it the hard way) the joy has been sucked out of it a bit.
If you wish to make software engineering your job then it stops being a game where we ban things simply because they provide an “unfair” advantage.
If you wish to play at programming “the way it used to be done” then by all means go ahead. But mixing methapors between sports and work is not helpful.
To me, the effort part only feels pointless if you want to accept LLM-derived code as comparable. Any marathon run is pointless, you go in a loop and we have cars. The point is to do the run yourself unaided.
That a computer can play a virtual instrument better than me (and continuously better than anyone) has little bearing on whether or not I continue practicing on my piano.
But software engineer, as a job rather than a hobby, has always been about results. Hence my original point.
Hence my confusion. I am not perusing my hobbies for “the world at large” I’m perusing them for myself; being able to drive 26 miles makes no difference to me nor to the others who wish to run it.
People who continue to do the hobby without LLMs will still get unique skills, like woodworkers who don’t use power tools.
I understand that this is a loss for those who were pursuing programming as a hobby only as long as it gave them a unique skill and made them feel superior to others. However, for better or worse, I can’t bring myself to care about that.
Hobby programming without LLMs and you gain unique skills (you can do the work without power tools).
Hobby program with LLMs, you gain unique skills again (you learn how to properly use these models and get to quickly iterate on your design skills). Regardless of what the big labs are saying, there is still a lot of skill involved in creating with them; if only because they don’t read minds so you have the get good at describing your taste and requirements.
This is pretty spot on. But it's important to note that the resulting code, is still often significantly better than what those developers used to produce. Especially under tight deadlines and pressure to complete CRUD like interfaces. We simply can't afford spending a whole day tweaking a sortable table on the web anymore.
But then again, LLMs do offer benefits that are impossible not to want to harness.
- Why should I even start vibing up my desired tools and projects when I can just wait it out so a better clanker emerges 6 months down the line?
- What guarantee is there that Rust is the best programming language for vibe coding when the models could improve to directly output assembly? What incentive do you have in vibing up something?
An agentic workflow lets me literally sit on my hands (or play Star Wars: Galactic Racer, or play my musical instrument, or work on my CAD hobby, etc) while the agent machine does my work for me.
The best part is that this is how my employer wants me to work. They literally want me to be lazy. Works for me!
Where is my incentive structure to not use AI?
[1] coding manually brings me a non-zero amount of joy and mental satisfaction and stimulation but I am still trading time for money.
Generally, automation does not make us work less because the expectations rise equivalently.
2. I’m paid to be a warm body who is available. That’s I am paid salaried instead of hourly.
3. I don’t really care if this situation changes in the future. If AI takes my job there’s nothing I can personally do to stop that from happening. I’m just saving my money and planning to retire as early as possible. I view work as a necessary evil.
To answer your question: 1) do you want it now, or in six months? and 2) it will still take the same amount of human effort.
The environmental aspects can be solved.
Quality is a measure of the person driving the process, models can write code better than the average engineer.
The impact on learning, seniority etc is the one big unknown - but I believe can/will also be solved.
If you're not interested in "average", volume isn't going to make up for quality. I'm certainly not perfectly satisfied with LLM output quality, but even the best programmer in the world will undeniably be helped by LLM tooling, even if only for having it write tests, benchmarks, run experiments, create research summaries, and so on.
I don't think that matters. What matters is that they ARE solved. Which is not the case today.
I’m sure the people dying to air pollution and high temperatures feel good knowing that it can hypothetically be solved.
I work for a FAANG, we have unlimited AI credits to be used on any model we want, with basically any tooling we want.
My whole team massively uses AI, I barely do. I use not even 10% of the tokens that they do. Yet I have not only the highest rate of PRs raised, but also the highest rate of PRs I reviewed.
AI is all about fake productivity. You feel productive while you're achieving the same, if not less.
Which seems like the right way to use tools?
1. OP probably works on a team that does not ship much. It's a dead end team on maintenance mode.
2. OPs colleagues were not good engineers even before LLMS
3. OP is lying.
4. OP is delusional and is actually left behind. And is measuring wrong metrics.
Which one did I get correct and wrong? Does you manager also agree you are the most productive engineer in team?
A good software engineer has always been able to write a ton of code all by themselves in a short period of time without any oversight. There's nothing romantic or intelligent about that. It's just reckless.
Teamwork is more important than ever, and yes of course that slows things down. That's the point. The decision making is the slowest part. 5 hours of meetings today can completely change the trajectory of the next 5 years. This isn't a revelation for anyone but the most naive.
We now have people commonly running multiple concurrent programming tasks, while automatically and interactively testing them, reviewing them, and validating them. They're then handed for human review, where little is found because the quality regime is now set so high. We're talking weeks of development + testing/validating time brought down into the realm of days.
So this is pure "my horse & cart are just as good as the motorcar"-levels of argument. If I bury my head in the sand long enough, and deep enough, this will all just blow over, and we'll go back to doing stuff significantly slower/worse any day now...
The quality of stuff I am shipping today is the highest it has ever been, and tested/validated significantly more comprehensively. Why? Because time/resources was the limitation, not willingness or skill. I always wanted to test for every single possible edge case/scenario, but it wasn't in the time-budget (and diminish returns). Now I don't need to make that trade-off.
At work I still read through every line in the diff (unless it's a personal util/non-production facing/critical etc). Even with this I feel less 'in touch' with my work code than I do with my home code.
LLMs are tools like motor vehicles. They allow you to get somewhere, quickly but brutishly. You won't be able to appreciate the beautiful scenery much, as it rushes by. You're going to have to compensate for the exercise you would have gotten, had you walked, or it's going to be unhealthy. It's undeniably bad for the environment (electric or not) and has other negative externalities. The politics of many of the countries supplying most motor vehicle fuel to the world are beyond fascist by modern leftist standards. Yet, we depend on them, we have constructed a lot of the infrastructure we need to live around their existence, and even if every city became walkable over night, the last mile for most transport would be covered by some motor vehicle. There are many "what ifs" surrounding motor vehicles where the answer boils down to: "Well, then we're fucked". But we're not fucked today, and so on it goes.
Of course you can try not to partake in motor vehicles, and insofar as you don't make a moral grandstand about it, that's totally fine with me. But the rest of the world is in it for the ride. The first cars weren't very good either.
If you're afraid of being called a luddite, you've already lost.
You shouldn't value the opinion of those who would insult you for not wanting to offload something you find intellectually fulfilling to a machine.
You can still manually type code, it can be a great hobby, but it is unlikely that someone will pay you a full time salary for doing it