> it could also be that these software jobs won’t pay as well as in the past, because, of course, the jobs aren’t as hard as they used to be. Acquiring the skills isn’t as challenging.
This sounds opposite to what the article said earlier: newbies aren’t able to get as much use out of these coding agents as the more experienced programmers do.
NYT has it out for digital advertisers, who directly compete with them. I do sense some schadenfreude here that the tech nerds who work at these places might be in trouble.
"Silicon Valley panjandrums spent the 2010s lecturing American workers in dying industries that they needed to “learn to code."
To copywriters at the NYT, LLMs are far better at stringing together natural language prose than large amounts of valid software. Get ready to supervise LLMs all day if you're not already.
Because we love tech? I'm absolutely terrified about the future of employment in this field, but I wouldn't give up this insane leap of science fiction technology for anything.
A really good pattern-matching engine is an "insane leap of science fiction"? It saves me a bit of typing here and there with some good pattern matching. Trying to get it to do anything more than a few lines gives me gibberish, or an infinite loop of "Oh, you're right, I need to do X, not Y", over and over - and that's Opus 4.5 or whatever the recent one is.
Would you give it access to your bank account, your 401k, trust it to sell your house, etc? I sure wouldn't.
>A really good pattern-matching engine is an "insane leap of science fiction"?
Yes, literally. The ship computer voice interface in Star Trek was complete science fiction until 2022. Now its ability to understand speech and respond seem quaint in comparison to current AI.
What is a coder? Someone who is handed the full specs and sits down and just types code? I have never met such a person. The most annoying part of SWE is everyone who isn't an SWE has inane ideas about what we do.
>A.I. had become so good at writing code that Ebert, initially cautious, began letting it do more and more. Now Claude Code does the bulk of it.
is a little overstated. I think the brownfield section has things exactly backwards. Claude Code benefits enormously from large, established codebases, and it’s basically free riding on the years of human work that went into those codebases. I prodded Claude to add SNFG depictions to the molecular modeling program I work on. It couldn’t have come up with the whole program on its own and if I tried it would produce a different, maybe worse architecture than our atomic library, and then its design choices for molecules might constrain its ability to solve the problem as elegantly as it did. Even then, it needed a coworker to tell me that it had used the incorrect data structure and needed to switch to something that could, when selected, stand in for the atoms it represented.
Also this:
>But A.I.-generated code? If it passes its tests and works, it’s worth as much as what humans get paid $200,000 or more a year to compose.
Isn’t really true. It’s the free-riding problem again. The thing about an ESP is that the LLM has the advantage of either a blank canvas (if you’re using one to vibe code a startup), or at least the fact that several possibilities converge on one output, but, genuinely, not all of those realities include good coding architecture. Models can make mistakes, and without a human in the loop those mistakes can render a codebase unmaintainable. It’s a balance. That’s why I don’t let Claude stamp himself to my commits even if he assisted or even did all the work. Who cares if Claude wrote it? I’m the one taking responsibility for it. The article presents Greenfield as good for a startup, and it might be, but only for the early, fast, funding rounds, when you have to get an MVP out right now. That’s an unstable foundation they will have to go back and fix for regulatory or maintenance reasons, and I think that’s the better understanding of the situation than framing Aayush’s experience as a user error.
Even so, “weirdly jazzed about their new powers” is an understatement. Every team including ours has decades of programmer-years of tasks in the backlog, what’s not to love about something you can set to pet peeves for free and then see if the reality matches the ideal? git reset --hard if you don't like what it does, and if you do all the better. The Cuisy thing with the script for the printer is a perfect application of LLMs, a one-off that doesn’t have to be maintained.
Also, the whole framing is weirdly self limiting. The architectural taste that LLMs are, again, free riding off of, is hard won by doing the work more senior engineers are giving to LLMs instead of juniors. We’re setting ourselves up for a serious coordinated action problem as a profession. The article gestures at this a couple times
The thing about threatening LLMs is pretty funny too but something in me wants to fall back to Kant's position that what you do to anything you do to yourself.
Another trash article from the New York Times, who financially benefit from this type of content because of their ongoing litigation against OpenAI. I think the assumption that developers don't code is wrong. Most software engineers don't even want to code, they are opportunists looking to make money. I have yet to experience this cliff of coding. These people aren't asking for hard enough questions. I have a bunch of things I want AI to build that it completely fails on.
The article could have been written from a very different perspective. Instead, the "journalists" likely interviewed a few insiders from Big Tech and generalized. They don't get it. They never will.
Before the advent of ChatGPT, maybe 2 in 100 people could code. I was actually hoping AI would increase programming literacy but it didn't, it became even more rare. Many journalists could have come at it from this perspective, but instead painted doom and gloom for coders and computer programming.
The New York Times should look in the mirror. With the advent of the iPad, most experts agreed that they would go out of business because a majority of their revenue came from print media. Look what happened.
Understand this, most professional software and IT engineers hate coding. It was a flex to say you no longer code professionally before ChatGPT. It's still a flex now. But it's corrupt journalism when there is a clear conflict of interest because the NYT is suing the hell out of AI companies.
I agree that the article is a poor take on AI in programming. However, I wouldn't blame NYT for corrupt journalism. This is an op-ed, not something written by NYT staff.
You have to hold AI hand to do even simple vanilla JS correctly. Or do framework code which is well documented all over the net. I love AI and use it for programming a lot, but the limitations are real.
Most of this thread is debating whether models are good or bad at writing code... however, I think a more important question is what we feed the AI with because that dramatically determines the quality of the output.
When your agent explores your codebase trying to understand what to build, it read schema files, existing routes, UI components etc... easily 50-100k tokens of implementation detail. It's basically reverse-engineering intent from code. With that level of ambiguous input, no wonder the results feel like junior work.
When you hand it a structured spec instead including data model, API contracts, architecture constraints etc., the agent gets 3-5x less context at much higher signal density. Instead of guessing from what was built it knows exactly what to build. Code quality improves significantly.
I've measured this across ~47 features in a production codebase with amedian ratio: 4x less context with specs vs. random agent code exploration. For UI-heavy features it's 8-25x. The agent reads 2-3 focused markdown files instead of grepping through hundreds of KB of components.
To pick up @wek's point about planning from above: devs who get great results from agentic development aren't better prompt engineers... they're better architects. They write the spec before the code, which is what good engineering always was... AI just made the payoff for that discipline 10x more visible.
The other day I (well, the AI) just wrote a Rust app to merge two (huge, GB of data) tables by discovering columns with data in common based on text distance (levenshtein and Dice) . It worked beautifully
An i have NEVER made one line of Rust.
I dont understand nay-sayers, to me the state of gen.AI is like the simpsons quote "worst day so far". Look were we are within 5 years of the first real GPT/LLM. The next 5 years are going to be crazy exciting.
The "programmer" position will become a "builder". When we've got LLMs that generate Opus quality text at 100x speed (think, ASIC based models) , things will get crazy.
"Can you believe that Dad actually used to have to go into an office and type code all day long, MAUALLY??! Line by line, with no advice from AI, he had to think all by himself!"
I was thinking about that recently. Maybe decades from now people will look at things like the Linux kernel or Doom and be shocked that mere humans were able to program large codebases by hand.
This was literally part of the premise of The Jetsons. George's job was to press a single button while the computer RUDI did all the work.
The difference is, Jetsons wasn't a dystopia (unlike the current timeline), so when Mr. Spacely fired George, RUDI would take his side and refuse to work until George was re-hired.
It's really time that mainstream media picks up on 'agentic coding' and the implications of writing software becoming a commodity.
I'm an engineer (not only software) by heart, but after seeing what Opus 4.6 based agents are capable of and especially the rate of improvement, i think the direction is clear.
I keep getting stuck on the liability problem of this supposed "new world". If we take this as far as it goes: AI agent societies that designs, architects, and maintains the entire stack E2E with little to no oversight. What happens when rogue AIs do bad things? Who is responsible? You have to have fireable senior engineers that understand deep fundamentals to make sure things aren't going awry, right? /s
Having an AI is like having a dedicated assistant or junior programmer that sometimes has senior-level insights. I use it to do tedious tasks where I don't care about the code - like today I used it to generate a static web page that let me experiment with the spring-ai chat bot code I was writing - basic. But yesterday it was able to track down the cause of a very obscure bug having to do with a pom.xml loading two versions of the same library - in my experience I've spent a full day on that type of bug and Claud was able to figure it out from the exception in just minutes.
But when I've used AI to generate new code for features I care about and will need to maintain it's never gotten it right. I can do it myself in less code and cleaner. It reminds me of code in the 2000s that you would get from your team in India - lots of unnecessary code copy-pasted from other projects/customers (I remember getting code for an Audi project that had method names related to McDonalds)
I think though that the day is coming where I can trust the code it produces and at that point I'll just by writing specs. It's not there yet though.
I've generated 250KLoC this week, absolutely no changes in deps or any other shenanigans. I'm not even really trying to optimize my output. I work on plans/proposals with 2 or 3 agents simultaneously in Cursor while one does work, sometimes parallelized. I can't do that in less code and cleaner. I can't do it at all. Don't wait too long.
Syntax highlighting rules, initially vibe-coded 40 languages and formats in about 10 minutes. What surprised me is when it switched the design from a class to the far more elegant single line of code:
>I think though that the day is coming where I can trust the code it produces and at that point I'll just by writing specs. It's not there yet though.
Must be nice to still have that choice. At the company I work for they've just announced they're cancelling all subscriptions to JetBrains, Visual Studio, Windsurf, etc. and forcing every engineer to use Claude Code as a cost-saving measure. We've been told we should be writing prompts for Claude instead of working in IDEs now.
I'm halfway through Steve Yegge's book Vibe Coding. Yegge was quoted in the article:
> “We’re talking 10 to 20 — to even 100 — times as productive as I’ve ever been in my career,” Steve Yegge, a veteran coder who built his own tool for running swarms of coding agents
That tool has been pretty popular. It was a couple hundred thousand lines of code and he wrote it in a couple months. His book is about using AI to write major new projects and get them reliable and production-ready, with clean, readable code.
It's basically a big dose of solid software engineering practices, along with enough practice to get a feel for when the AI is screwing up. He said it takes about a year to get really good at it.
(Yegge, fwiw, was a lead dev at Amazon and Google, and a well-known blogger since the early 2000s.)
It's an accelerator. A great tool if used well. But just like all the innovations before it that were going to replace programmers it simply won't.
I used Claude just the other day to write unit test coverage for a tricky system that handles resolving updates into a consistent view of the world and handles record resurrection/deletion. It wrote great test coverage because it parsed my headerdoc and code comments that went into great detail about the expected behavior. The hard part of that implementation was the prose I wrote and the thinking required to come up with it. The actual lines of code were already a small part of the problem space. So yeah Claude saved me a day or two of monotonously writing up test cases. That's great.
Of course Claude also spat out some absolute garbage code using reflection to poke at internal properties because the access level didn't allow the test to poke at the things it wanted to poke at, along with some methods that were calling themselves in infinite recursion. Oh and a bunch of lines that didn't even compile.
The thing is about those errors: most of them were a fundamental inability to reason. They were technically correct in a sense. I can see how a model that learned from other code written by humans would learn those patterns and apply them. In some contexts they would be best-practice or even required. But the model can't reason. It has no executive function.
I think that is part of what makes these models both amazingly capable and incredibly stupid at the same time.
For one thing comments here appear to apply to the quality and issues today not potentially going forward. Quality will change quicker than anyone expects. I am wondering how many people at HN remember when the first Mac came out with Mac Paint and then Pagemaker or Quark. That didn't evolve anywhere nearly as quickly as AI appears to be.
Also I am not seeing how anyone is considering that what a programmer considers quality and what 'gets the job done' (as mentioned in the article) matters in any business. (Example with typesetting is original laser printers were only 300dpi but after a short period became 1200dpi 'good enough' for camera ready copy).
> “The reason that tech generally — and coders in particular — see L.L.M.s differently than everyone else is that in the creative disciplines, L.L.M.s take away the most soulful human parts of the work and leave the drudgery to you,” Dash says. “And in coding, L.L.M.s take away the drudgery and leave the human, soulful parts to you.”
This doesn’t really make sense to me. GenAI ostensibly removes the drudgery from other creative endeavors too. You don’t need to make every painstaking brushstroke anymore; you can get to your intended final product faster than ever. I think a common misunderstanding is that the drudgery is really inseparable from the soulful part.
Also, I think GenAI in coding actually has the exact same failure modes as GenAI in painting, music, art, writing, etc. The output lacks depth, it lacks context, and it lacks an understanding of its own purpose. For most people, it’s much easier to intuitively see those shortcomings of GenAI manifest in traditional creative mediums, just because they come more naturally to us. For coding, I suspect the same shortcomings apply, they just aren’t as clear.
I mean, at the end of the day if writing code is just to get something that works, then sure, let’s blitz away with LLMs and not bother to understand what we’re doing or why we do it anymore. Maybe I’m naive in thinking that coding has creative value that we’re now throwing away, possibly forever.
Maybe they mean more soulful like a fellow that blacksmiths his own tools and metal fasteners prior to constructing something. I’d personally think this person was a badass, but until wwiii, it’s so impractical and seems arbitrary because why stop there - get more soulful and mine your own ore too.
It's all nonsense. It's just better search, intelligence in not artificial. They are trying to convince everyone that they don't need to pay programmers. That's all, all it is. It'll work on the ignorant who'll take less money to make sure it works and fix the bugs, which is mostly what they were paying for anyway. They just want to devalue the work of the people they are reliant on. Nothing new.
how many times in the history of computer programming has there been an end to computer programming as we know it, successfully, and how many times predicted?
I can think of one successfully, off hand, although you could probably convince me there was more than one.
the principle phrase being "as we know it", since that implies a large scale change to how it works but it continues afterwards, altered.
If coding truly becomes effortless to produce - and by that extension a product becomes near free to produce - then I find it quite odd that the executive class thinks their businesses won’t be completely up ended by a raging sea of competition.
I'm not normally a fan of the NYT but this wasn't too bad. It passed the Gel-Mann test, and is clearly written by someone who knows the field well, even though the selection of quotes skews to towards outliers -- I think Yeggie for instance is pretty far out of the mainstream in his views on LLMs, whether ahead or sideways.
As a result a lot of the responses here are either quibbles or cope disguised as personal anecdotes. I'm pretty worried about the impact of the LLMs too, but if you're not getting use out of them while coding, I really do think the problem is you.
Since people always want examples, I'll link to a PR in my current hobby project, which Claude code helped me complete in days instead of weeks. https://github.com/igor47/csheet/pull/68 Though this PR creates a bunch of tables, routes, services -- it's not just greenfield CRUD work. We're figuring out how to model a complicated domain (the rules to DnD 5e, including the 2014 and the 2024 revisions of those rules), integrating with existing code, thinking through complex integrations including with LLMs at run time. Claude is writing almost all the code, I'm just steering
I've found LLMs to be extremely powerful in research projects. A lot of code related to research is very bespoke by its nature. Using Codex, I've been able to iterate on ideas that I would never had the time or courage to explore before. As for code quality/brevity, it doesn't really matter in this context as long as it works. And it's incredible to have this companion that understands broadly every aspect of tangential knowledge required to execute an idea. I do think it helps that I have over 25 years of experience in my domain (geospatial), which helps me guardrail my interactions and get good results in as few shots as possible.
It’s hard for me to believe that, unless you’re just doing simple glue work or you’re working in a low stakes environments, anyone is just delegating everything to agents. If you’re working on a
migration (common in enterprise infrastructure work), you’re familiar with the needless abstractions, and it’s something you’ve done many times over; agents can certainly expedite change. If you’re building anything with depth and you do not have a clear understanding of the underpinning logic, you’re either very gifted in your ability to reason about abstractions or you’re setting yourself for a failure at some point in the future. You need expertise at some point. Programming/debugging as a means of learning a domain is akin to writing as a means of clarifying your thoughts.
That being said, yea enterprise coding can be extremely mundane and it’s setup for learning it deeply then finding a way to do it faster. I’m likely in the 90% range of my work being done by Claude, but I’m working in a domain I’ve got years of experience with hand coding and stepping through code in my debugger.
I think this latter piece is the challenge I’m struggling with. There is an endless amount of work that can be done at my company but as long as the economy is in a weird spot, I’m being led to believe that ai is making me expendable. This is a consequence of the fact that glue work represents 80% of my output (not value). The other 20% of time at work is exploring ideas without guaranteed results, its aligning stakeholders, its testing feasibility with mvps or experts from another area I need some help with. If glue work represents tangible output and conceptual work is something that may not actually have value my manager wants me to explore it, I’m just a glue guy in enterprise while I’m left chasing the dragon of a cool project for me to really sink my teeth into. That project is just a half baked bad idea from someone disconnected with reality. Glue work is measurable in LoC (however useless a metric it is measurable) and it’s certainly paying the bills.
91 comments
[ 3.3 ms ] story [ 100 ms ] threadThis sounds opposite to what the article said earlier: newbies aren’t able to get as much use out of these coding agents as the more experienced programmers do.
"Silicon Valley panjandrums spent the 2010s lecturing American workers in dying industries that they needed to “learn to code."
To copywriters at the NYT, LLMs are far better at stringing together natural language prose than large amounts of valid software. Get ready to supervise LLMs all day if you're not already.
The brain rot from the author couldn't even think of "unit test".
Would you give it access to your bank account, your 401k, trust it to sell your house, etc? I sure wouldn't.
Yes, literally. The ship computer voice interface in Star Trek was complete science fiction until 2022. Now its ability to understand speech and respond seem quaint in comparison to current AI.
I’ve tended to hold the same opinion of what the average SWE thinks everyone else does.
This excerpt:
>A.I. had become so good at writing code that Ebert, initially cautious, began letting it do more and more. Now Claude Code does the bulk of it.
is a little overstated. I think the brownfield section has things exactly backwards. Claude Code benefits enormously from large, established codebases, and it’s basically free riding on the years of human work that went into those codebases. I prodded Claude to add SNFG depictions to the molecular modeling program I work on. It couldn’t have come up with the whole program on its own and if I tried it would produce a different, maybe worse architecture than our atomic library, and then its design choices for molecules might constrain its ability to solve the problem as elegantly as it did. Even then, it needed a coworker to tell me that it had used the incorrect data structure and needed to switch to something that could, when selected, stand in for the atoms it represented.
Also this:
>But A.I.-generated code? If it passes its tests and works, it’s worth as much as what humans get paid $200,000 or more a year to compose.
Isn’t really true. It’s the free-riding problem again. The thing about an ESP is that the LLM has the advantage of either a blank canvas (if you’re using one to vibe code a startup), or at least the fact that several possibilities converge on one output, but, genuinely, not all of those realities include good coding architecture. Models can make mistakes, and without a human in the loop those mistakes can render a codebase unmaintainable. It’s a balance. That’s why I don’t let Claude stamp himself to my commits even if he assisted or even did all the work. Who cares if Claude wrote it? I’m the one taking responsibility for it. The article presents Greenfield as good for a startup, and it might be, but only for the early, fast, funding rounds, when you have to get an MVP out right now. That’s an unstable foundation they will have to go back and fix for regulatory or maintenance reasons, and I think that’s the better understanding of the situation than framing Aayush’s experience as a user error.
Even so, “weirdly jazzed about their new powers” is an understatement. Every team including ours has decades of programmer-years of tasks in the backlog, what’s not to love about something you can set to pet peeves for free and then see if the reality matches the ideal? git reset --hard if you don't like what it does, and if you do all the better. The Cuisy thing with the script for the printer is a perfect application of LLMs, a one-off that doesn’t have to be maintained.
Also, the whole framing is weirdly self limiting. The architectural taste that LLMs are, again, free riding off of, is hard won by doing the work more senior engineers are giving to LLMs instead of juniors. We’re setting ourselves up for a serious coordinated action problem as a profession. The article gestures at this a couple times
The thing about threatening LLMs is pretty funny too but something in me wants to fall back to Kant's position that what you do to anything you do to yourself.
The article could have been written from a very different perspective. Instead, the "journalists" likely interviewed a few insiders from Big Tech and generalized. They don't get it. They never will.
Before the advent of ChatGPT, maybe 2 in 100 people could code. I was actually hoping AI would increase programming literacy but it didn't, it became even more rare. Many journalists could have come at it from this perspective, but instead painted doom and gloom for coders and computer programming.
The New York Times should look in the mirror. With the advent of the iPad, most experts agreed that they would go out of business because a majority of their revenue came from print media. Look what happened.
Understand this, most professional software and IT engineers hate coding. It was a flex to say you no longer code professionally before ChatGPT. It's still a flex now. But it's corrupt journalism when there is a clear conflict of interest because the NYT is suing the hell out of AI companies.
When your agent explores your codebase trying to understand what to build, it read schema files, existing routes, UI components etc... easily 50-100k tokens of implementation detail. It's basically reverse-engineering intent from code. With that level of ambiguous input, no wonder the results feel like junior work.
When you hand it a structured spec instead including data model, API contracts, architecture constraints etc., the agent gets 3-5x less context at much higher signal density. Instead of guessing from what was built it knows exactly what to build. Code quality improves significantly.
I've measured this across ~47 features in a production codebase with amedian ratio: 4x less context with specs vs. random agent code exploration. For UI-heavy features it's 8-25x. The agent reads 2-3 focused markdown files instead of grepping through hundreds of KB of components.
To pick up @wek's point about planning from above: devs who get great results from agentic development aren't better prompt engineers... they're better architects. They write the spec before the code, which is what good engineering always was... AI just made the payoff for that discipline 10x more visible.
An i have NEVER made one line of Rust.
I dont understand nay-sayers, to me the state of gen.AI is like the simpsons quote "worst day so far". Look were we are within 5 years of the first real GPT/LLM. The next 5 years are going to be crazy exciting.
The "programmer" position will become a "builder". When we've got LLMs that generate Opus quality text at 100x speed (think, ASIC based models) , things will get crazy.
Where's the references to the decline in quality and embarrassing outages for Amazon, Microsoft, etc?
"Can you believe that Dad actually used to have to go into an office and type code all day long, MAUALLY??! Line by line, with no advice from AI, he had to think all by himself!"
"Dad, I've sent out 1000 applications and haven't had a call back. I can't take it anymore. Has it always been like this?"
The Dad: It's not my fault!
The difference is, Jetsons wasn't a dystopia (unlike the current timeline), so when Mr. Spacely fired George, RUDI would take his side and refuse to work until George was re-hired.
I'm an engineer (not only software) by heart, but after seeing what Opus 4.6 based agents are capable of and especially the rate of improvement, i think the direction is clear.
Citation needed. Are most developers "rarely" writing code?
But when I've used AI to generate new code for features I care about and will need to maintain it's never gotten it right. I can do it myself in less code and cleaner. It reminds me of code in the 2000s that you would get from your team in India - lots of unnecessary code copy-pasted from other projects/customers (I remember getting code for an Audi project that had method names related to McDonalds)
I think though that the day is coming where I can trust the code it produces and at that point I'll just by writing specs. It's not there yet though.
Wire up authentication system with sso. done Setup websockets, stream audio from mic, transcribe with elvenlabs. done.
Shit that would take me hours takes literally 5 mins.
The rules files:
* https://repo.autonoma.ca/repo/treetrek/tree/HEAD/render/rule...
Must be nice to still have that choice. At the company I work for they've just announced they're cancelling all subscriptions to JetBrains, Visual Studio, Windsurf, etc. and forcing every engineer to use Claude Code as a cost-saving measure. We've been told we should be writing prompts for Claude instead of working in IDEs now.
> “We’re talking 10 to 20 — to even 100 — times as productive as I’ve ever been in my career,” Steve Yegge, a veteran coder who built his own tool for running swarms of coding agents
That tool has been pretty popular. It was a couple hundred thousand lines of code and he wrote it in a couple months. His book is about using AI to write major new projects and get them reliable and production-ready, with clean, readable code.
It's basically a big dose of solid software engineering practices, along with enough practice to get a feel for when the AI is screwing up. He said it takes about a year to get really good at it.
(Yegge, fwiw, was a lead dev at Amazon and Google, and a well-known blogger since the early 2000s.)
Just checking that you're using maven-enforcer-plugin
I used Claude just the other day to write unit test coverage for a tricky system that handles resolving updates into a consistent view of the world and handles record resurrection/deletion. It wrote great test coverage because it parsed my headerdoc and code comments that went into great detail about the expected behavior. The hard part of that implementation was the prose I wrote and the thinking required to come up with it. The actual lines of code were already a small part of the problem space. So yeah Claude saved me a day or two of monotonously writing up test cases. That's great.
Of course Claude also spat out some absolute garbage code using reflection to poke at internal properties because the access level didn't allow the test to poke at the things it wanted to poke at, along with some methods that were calling themselves in infinite recursion. Oh and a bunch of lines that didn't even compile.
The thing is about those errors: most of them were a fundamental inability to reason. They were technically correct in a sense. I can see how a model that learned from other code written by humans would learn those patterns and apply them. In some contexts they would be best-practice or even required. But the model can't reason. It has no executive function.
I think that is part of what makes these models both amazingly capable and incredibly stupid at the same time.
Also I am not seeing how anyone is considering that what a programmer considers quality and what 'gets the job done' (as mentioned in the article) matters in any business. (Example with typesetting is original laser printers were only 300dpi but after a short period became 1200dpi 'good enough' for camera ready copy).
This doesn’t really make sense to me. GenAI ostensibly removes the drudgery from other creative endeavors too. You don’t need to make every painstaking brushstroke anymore; you can get to your intended final product faster than ever. I think a common misunderstanding is that the drudgery is really inseparable from the soulful part.
Also, I think GenAI in coding actually has the exact same failure modes as GenAI in painting, music, art, writing, etc. The output lacks depth, it lacks context, and it lacks an understanding of its own purpose. For most people, it’s much easier to intuitively see those shortcomings of GenAI manifest in traditional creative mediums, just because they come more naturally to us. For coding, I suspect the same shortcomings apply, they just aren’t as clear.
I mean, at the end of the day if writing code is just to get something that works, then sure, let’s blitz away with LLMs and not bother to understand what we’re doing or why we do it anymore. Maybe I’m naive in thinking that coding has creative value that we’re now throwing away, possibly forever.
I can think of one successfully, off hand, although you could probably convince me there was more than one.
the principle phrase being "as we know it", since that implies a large scale change to how it works but it continues afterwards, altered.
Not unlike all the A.I. companies all determined to build the machine god while predicting it’ll be disastrous. Same thing - better it starts with us
As a result a lot of the responses here are either quibbles or cope disguised as personal anecdotes. I'm pretty worried about the impact of the LLMs too, but if you're not getting use out of them while coding, I really do think the problem is you.
Since people always want examples, I'll link to a PR in my current hobby project, which Claude code helped me complete in days instead of weeks. https://github.com/igor47/csheet/pull/68 Though this PR creates a bunch of tables, routes, services -- it's not just greenfield CRUD work. We're figuring out how to model a complicated domain (the rules to DnD 5e, including the 2014 and the 2024 revisions of those rules), integrating with existing code, thinking through complex integrations including with LLMs at run time. Claude is writing almost all the code, I'm just steering
I've always hated solving puzzles with my deterministic toolbox, learning along the way and producing something of value at the end.
Glad that's finally over so I can focus on the soulful art of micromanaging chatbots with markdown instead.
That being said, yea enterprise coding can be extremely mundane and it’s setup for learning it deeply then finding a way to do it faster. I’m likely in the 90% range of my work being done by Claude, but I’m working in a domain I’ve got years of experience with hand coding and stepping through code in my debugger.
I think this latter piece is the challenge I’m struggling with. There is an endless amount of work that can be done at my company but as long as the economy is in a weird spot, I’m being led to believe that ai is making me expendable. This is a consequence of the fact that glue work represents 80% of my output (not value). The other 20% of time at work is exploring ideas without guaranteed results, its aligning stakeholders, its testing feasibility with mvps or experts from another area I need some help with. If glue work represents tangible output and conceptual work is something that may not actually have value my manager wants me to explore it, I’m just a glue guy in enterprise while I’m left chasing the dragon of a cool project for me to really sink my teeth into. That project is just a half baked bad idea from someone disconnected with reality. Glue work is measurable in LoC (however useless a metric it is measurable) and it’s certainly paying the bills.