"Fourth, we will not find ourselves working less. Rather, there will be a bimodal outcome: those unable to find a place in the new AI-powered industry will fall out of it entirely, while those that remain are worked harder and harder as they drive automated development systems. Five PRs a week? Hah! Try fifty. Or five hundred."
My experience with AI/LLMs summed up. The baseline expectation got higher. I didn't get any time back. My life didn't become easier.
As it was the case with all productivity catalysts in the past. The Industrial Revolution promised machines would work and we’d get to have leisure time. Instead, it brought 16 hour factory shifts. This will be no different
To paraphrase Marx, "And so, the equivalent amount of labour required to produce the goods needed to sustain a worker for a day is, say, 4 hours. But that doesn't mean the selfsame worker can work for no longer that 4 hours! He can be forced to work six, or eight, or twelve hours a day, and whatever additional goods he produces — that's the surplus product, which in this case goes straight to line up the factory owner's pockets".
Many economists, already from the 19th century, claimed that technological progress would reduce the workweek. Some imagined we'd work 3 days per week, others only 1. Some imagined we'd work a few hours everyday.
I remember many people even on this very site claiming AI would help humanity. I think the most ridiculous the most ridiculous claim was helping fight climate change, but helping produce more leisure time by automating work was definitely what some people thought, or at least what they wanted us to think while pushing their crap.
Literally no one, no one even claimed we would work less due to AI. It is one of many absurd claims about AI here. There was no period in which we would get think pieces predicting more leisure time.
We had plenty of think pieces about making people obsolete, about destruction, generally in celebratory tone. I cant even think of period where AI think pieces would promises much positive - it was sold to CEOs, so pitch was always "higher unemployment".
> Literally no one, no one even claimed we would work less due to AI. It is one of many absurd claims about AI here. There was no period in which we would get think pieces predicting more leisure time.
I think this is largely right. Software engineering as a stable, high-earning and in demand career is a thing of the past.
Those who make it work from here are those who've changed their work completely to resemble something more like a hands-off tech lead. There will of course be the lucky few who've made it work as artisans without direct commercial pressures to conform to the new "right way to do it". But the careers many of us have had is not a path available for others to follow anymore.
The irony of automating away "inefficiencies", "drudgery" and the "labour intensive" parts of other professions stings all the more when the sharp end of that shtick is pointed back in our direction.
Of course, on the other hand does it make sense to keep an obsolete field artificially alive because "jobs"? There are many new oportunities opening for us. One will be increasing demand for cybersecurity auditors, someone will need to wire those servers and someone else will need to expand the AI into more fields.
We humans don't like change much, but that doesn't mean we should pump money to blockbuster.
Quitting is an individual action with potential adverse consequences (misery).
The political question is why despite productivity in economic terms (which i know is flawed) growing many times over, do we still have to work as much, get paid so little, and have so many unemployed people looking for a job?
Looks like without a parasitic capitalist class, we could share resources and work and have people live better lives and work less.
Your job being automated to the point where you no longer proform it has the same adverse consequences. Whether this could work is some economic and societal setup is irrelevant, because that's not the world we live in.
What are you talking about? The only reason we could work less is because we always gave way to innovation. We don't haul large blocks of ice because "evil capitalists" started making fridges. Or do you want to reinstate ice trade, just because it generated more jobs? Or ban tractors because farming by hand was made by way more people?
"Sharing resources" doesn't work. We share way too much just to keep pensioners alive and can't have kids because of that. Confiscating all the billionaire's wealth wouldn't make a dent and would destroy much more.
> Third, once destroyed those industries will not meaningfully return. This is not a process that will result in a natural ebb and flow. Once the software development profession is disrupted, it will stay that way as talent pipelines dry up and expertise ages out
I can echo that. If/when I am replaced with an LLM, I plan on switching to a hands-on profession - welder, electrician, machinist. I am also not planning on ever going back to software... errr... prompt "engineering". Maybe for twice my pre-LLM salary at a place with a no-LLM policy? More at a prompt-engineering place.
They might very well go away. There is definitely an AI bubble, and it remains to be seen whether it deflates gradually or pops spectacularly. Geoeconomics might destroy them by constraining their access to hardware. The capabilities are real, but whether those capabilities are realised is a different matter.
What is China subsidises its inference providers to provide like utility level "intelligence" at or below cost globally until all the knowledge jobs in the west go the way of the manufacturing ones..
Providing services below cost is how SV and American tech operated last 25 years. It was the same pattern again and again - sell bellow cost, competition cant compete, create quasi monopoly, enshittify and start earning.
OpenAI and Anthropic still sells tokens under cost.
I was with the article - especially the bit about the 10% an LLM loses and how that loss can be de-amplified across eventual meaning, but then...
"My first instinct is to laugh and shake my head. One need not look very far to find indignant software developers absolutely certain that their jobs cannot possibly be automated away by the very tools their industry contemporaries are creating to replace them. I suspect you’d also not have to look far into their posting histories to find those same people comparing cabbies to buggy whip makers."
What a rude and callous comment. I'm one of those developers, I'd love to see an LLM even fractionally capable of some of the things my job entails. Laugh at me as I defend my lifelong career, why don't you? I'm also one who decries such things as the removal of local services, taxi and otherwise, for those in the cloud. Screw you, man.
> I'd love to see an LLM even fractionally capable of some of the things my job entails
If you don't mind could you please elaborate on what tasks or knowledge your job requires that you feel an LLM isn't even _fractionally_ capable of? I understand if you said that out of emotion and frustration, but if you're serious I'm intensely and genuinely curious because nowadays it seems like the frontier models are capable of more programming tasks than not with the proper harness and engineer. When was the last time you tried?
I just want to emphasize that this is not a further provocation, just potentially in awe of what you do at your job and would like to know more.
classic example of the goomba fallacy - "all software engineers are one single person-caricature I have in my head and are therefore a stupid walking contradiction whenever two of them express different opinions".
Do these hypothetical laid off software engineers with no empathy for buggy drivers exist outside the imagination of people who hate software engineers? I know in my education we had a bunch of mandatory courses about automation and the effects on workers, how to consult with workers on how to lift them up rather than be antagonistic, and consider different stakeholders (where "the business" or "product" was but one of many) etc.
It's not easy being a developer right now. We went from being dorky and unpopular in school to being superstars (at work) and now the writing's on the wall. I can sympathize, but I'll concede that our recently developed attention for the state of nature and the fair distribution of capital is .. interesting.
I don't disagree, but the absolute hubris to think that our job is particularly difficult is also ridiculous. One by one our responsibilities will be able to be shifted into highly leveragable software, whether it's years or decades away.
There’s no irony. Author made a straw man of some bad, bad developer laughing at poor cabbies and it all clicked once VCs came for the fellow professionals.
Tech bro lamenting and berating developers how immoral and hypocritical they’re for “laughing” (has this ever happened?) at other professions, while doing the same. Story as old as time.
The point of this sentence is to show that there are Software Developers who are looking at this in a short-sighted way. There doesn't seem anything rude about it.
Great piece, well written and succinctly sums up my thoughts.
The bit I still don’t understand is how we all put up with the hallucinations. I was questioning Gemini last night about whether it could analyse a Fourtet song and give me a break down of the structure from beginning to end. “Sure!” it said with the endless enthusiasm you get from Gen tools, and then proceeded to spit out an absolute sack of fabricated shit. I pushed back, it apologised, and then generated more crap that had nothing to do with reality, I pushed back, we looped again, still just total fiction: “the drums don’t come in until bar 16” on a song that opens with a drum loop, that kind of crap.
We’re so so far away from tools here that are anywhere near being trustworthy and accurate. And yet we (including myself) are chunking out code after code. It’s so bizarre.
I’m guessing it’s that humans don’t have capacity to deal with this kind of scenario - it’s like having a junior staff member who is utterly incredible 90% of the time - completely convincing in their certainty and skill level, and then 10% of the time you catch them doing a shit in their desk drawer because they couldn’t be arsed to walk to the toilet. AI’s are basically sociopaths.
Doesn’t the article make the argument that since you can write tests this is not as much of a problem for code gen ?
Its arguable whether it is a foolproof solution (I don’t think so) but it definitely makes it look like you can build a harness around the stochastic machine that will validate the correctness of the generated randomness.
Monkeys and typewriters when you can quickly validate whether it’s Shakespeare or not is a costly but theoretically feasible scenario. No?
> We’re so so far away from tools here that are anywhere near being trustworthy and accurate. And yet we (including myself) are chunking out code after code. It’s so bizarre.
I think one more thing this whole LLM charade in the last few years has revealed is that no-one really cares. As long as it "looks" like it works, turns out, its all fine.
Are people seriously getting use out of LLMs? Everything they produce is extermely sloppy in my experience, like actually mostly useless. I really dont understand the hype. Its very confusing.
I work 40 hours a week at a company I care very little for. Despite that, I still have to make 40 hours. I also have a side-project that I do in my free time. I do not have the energy to sit behind my desk for >40 hrs a week
So I walk, program, sit, get coffee, read a bit, come back and review the code. Most of the time it's fine, sometimes I had forgotten to mention something, and have to correct it, but this step doesn't take more than ~15 mins.
I then have a feature that would've taken me multiple days. Not because I need multiple days, but because I do not have the 8 hours of continuous time to work on it.
People have forgotten that when you start your programming day, you have to get up to speed which takes me longer than others. Let's say this takes ~15 mins. That means that if you spend 2 hours programming, ~12% of that was just getting back into the groove. LLMs do this instantly, it reads context, you can ask it questions, and you're up to speed again in less than a minute.
The point here is not that LLMs provide high quality code, but they do save you a bit of time and energy, which is worth a lot in my opinion.
A lot of inventions haven't been that ground-breaking; only there to save time. You can wash the dishes by hand, but you can also have a machine do it while you go watch TV.
I am convinced the people who swear by them either have very different work tasks from me, or they have very different ideas about what a job well done looks like.
I feel like everything I apply these things to sends me up a much messier and long winded route to a useable result, when compared to just doing it myself from the jump. Even the things they're ostensibly good at like sorting data comes out so messy it's practically net zero by the time you're done with quality control.
I tend to do the main structural parts myself and tell it to fill in the gaps and add tests, which sort of works 70% of the time. It may not be worth what my company is being charged.
Every time I have actually engaged in conversations with people making the same claim you do, it turned out they did not really invest time into learning how to work with coding agents: They assumed, given that they’re developers, the know how to code, and thus would knew how to steer a coding agent.
That is a wrong assumption, however. An agent is an entirely new tool in your toolbox, with no similarity to any of those you already have. You will need to learn how to wield it, like a new programming language or technology you’re unfamiliar with. You will need to do some small side projects to learn. You will need to develop a feeling for how it reacts to your inputs, when to reset the session, pass it links to documentation, or interrupt it.
None of this comes intuitively. It takes time and effort, and if you’re not ready to consciously invest that, coding agents are not going to work efficiently for you. That doesn’t take away from their utility though.
No, the more appropriate analogy would be that LLMs (reasoning agents would be better term, as they are no longer just an LLMs) are new beings rather than new tools.
A new tool can be ultimately understood, and has a well-defined behaviour you can rely on. Then you can indeed become an expert in using a new tool.
But AI agent is more a new being (or person). It's not possible to understand them, as they are not meant to. Each of them has individuality, and can change tomorrow. There is no guarantee of a common behaviour.
So you could replace "tool" in your comment with "person" and you will see the flaw of your argument. No matter how many people you saw before, it's difficult to generalize the skillset (at least for humans there are some biological and cultural arguments why they can be generalized over). You can always meet someone different, who's, for whatever reason, not being helpful to you.
Most of the interaction surface with a coding agent is the harness anyway, so I'd definitely classify them as tools. Moreover, I don't like the anthropomorphism of calling them beings—and even if I were to follow that narrative, you can absolutely become better at interacting with people by practicing it.
Regardless, it sure seems like developers experienced with using coding agents achieve better results than those without, which pretty much refuses your entire point.
Yes you can learn the harness, that's a tool. But it's not the important part.
I am not anthroporphizing them. I am just saying people are a better analogy.
> seems like developers experienced with using coding agents achieve better results than those without
That doesn't mean there is a skill involved. The same goes with people - you might just be naturally charismatic and get stuff done better with people. Doesn't mean it's a learnable skill that applies to everyone and every use case.
I don't know, it is pretty simple to type /init in claude and see if its understanding of a project matches reality. And if it can make changes to it.
The difference working with agents, is that a solo dev suddenly becomes a dev manager, who has to deal with a unreliable team of developers far out on the spectrum.
Without you telling what your job entails and how the pristine artifacts of your crystalline mind fit into the larger picture it's very hard to have a conversation about this.
But yes, many developers I know haven't coded manually in months and that includes me. That doesn't mean I drink coffee and take walks while the agent codes however. I'm now in the driving seat instead of wrangling syntax and waiting for my hands to type something.
I manage the rules, the intent, the structure. If I don't like what it does, I update the coding standard, the docs, the specs, you name it. If I want a different architecture I can actually get it done in 10min instead of a week - if at all. It's more effort than coding for me, because coding is slow and methodical and incentivizes pedantry, but for me it's a massive improvement. I have been bored out of my mind for a decade now, but I'm relatively good at what I do so I just stuck with it. I'm one of those "middle management types stuck in developer role"-type of guys so I guess my personality and proclivities have something to do with it. I also know quite a few grumpy old-school types that don't get anything done with LLMs. They can't communicate, they don't understand psychology, they don't understand architecture and have a debilitating case of missing the forest for the trees. The spread, in my experience, is massive.
Yes. Most of the people I see are on the extremes of either hating LLMs or fanatically loving it, but I am somewhere in the middle ground, I think, where I see it as a tool that sometimes helps to some extent. I don't depend on it, it almost never gives me new ideas or ways to do things, but for trivial tasks it helps me relax supervising it - usually writing the first draft for documentation or suggesting initial updates, even filling in some semi-repetitive code patterns in web apps ("create empty functions and link the buttons to it"). It is just a tool, not too bad to ignore, not too good to get more productivity, sometimes useful. Using it is like using an IDE 30 years ago or Intellisense 15-20 years ago.
LE. I see current versions of LLM like an intern that helps me doing work. We work together, I give directions and supervision and I am responsible for the results. I cannot give complex tasks and I cannot skip checking everything, but it usually helps.
I am getting serious use out of LLMs, and everyone else who does knows that those claiming otherwise are only getting slop because they’re not giving it enough guard rails, likely because they’re uninitiated, likely because they hate AI in the first place.
"Fourth, we will not find ourselves working less. Rather, there will be a bimodal outcome: those unable to find a place in the new AI-powered industry will fall out of it entirely, while those that remain are worked harder and harder as they drive automated development systems. Five PRs a week? Hah! Try fifty. Or five hundred."
My experience with AI/LLMs summed up. The baseline expectation got higher. I didn't get any time back. My life didn't become easier.
As it was the case with all productivity catalysts in the past. The Industrial Revolution promised machines would work and we’d get to have leisure time. Instead, it brought 16 hour factory shifts. This will be no different
> Instead, it brought 16 hour factory shifts.
To paraphrase Marx, "And so, the equivalent amount of labour required to produce the goods needed to sustain a worker for a day is, say, 4 hours. But that doesn't mean the selfsame worker can work for no longer that 4 hours! He can be forced to work six, or eight, or twelve hours a day, and whatever additional goods he produces — that's the surplus product, which in this case goes straight to line up the factory owner's pockets".
One of the big reasons why I hated AI since day one.
Why would it? Who, with a half brain, thought that you’d ever work less?
Many economists, already from the 19th century, claimed that technological progress would reduce the workweek. Some imagined we'd work 3 days per week, others only 1. Some imagined we'd work a few hours everyday.
I remember many people even on this very site claiming AI would help humanity. I think the most ridiculous the most ridiculous claim was helping fight climate change, but helping produce more leisure time by automating work was definitely what some people thought, or at least what they wanted us to think while pushing their crap.
Literally no one, no one even claimed we would work less due to AI. It is one of many absurd claims about AI here. There was no period in which we would get think pieces predicting more leisure time.
We had plenty of think pieces about making people obsolete, about destruction, generally in celebratory tone. I cant even think of period where AI think pieces would promises much positive - it was sold to CEOs, so pitch was always "higher unemployment".
> Literally no one, no one even claimed we would work less due to AI. It is one of many absurd claims about AI here. There was no period in which we would get think pieces predicting more leisure time.
https://www.newsbytesapp.com/news/business/ari-emanuels-3b-v...
https://www.telegraph.co.uk/business/2023/06/11/ignore-the-d...
https://time.com/6268804/artificial-intelligence-pissarides-...
> I cant even think of period where AI think pieces would promises much positive - it was sold to CEOs, so pitch was always "higher unemployment".
Higher unemployment means people work less.
I think this is largely right. Software engineering as a stable, high-earning and in demand career is a thing of the past.
Those who make it work from here are those who've changed their work completely to resemble something more like a hands-off tech lead. There will of course be the lucky few who've made it work as artisans without direct commercial pressures to conform to the new "right way to do it". But the careers many of us have had is not a path available for others to follow anymore.
The irony of automating away "inefficiencies", "drudgery" and the "labour intensive" parts of other professions stings all the more when the sharp end of that shtick is pointed back in our direction.
Fantastic article. However author never considered that I'm looking forward to my job being automated. That's like a whole point of programming.
To what end? Can't you just quit and get the same result?
Of course, on the other hand does it make sense to keep an obsolete field artificially alive because "jobs"? There are many new oportunities opening for us. One will be increasing demand for cybersecurity auditors, someone will need to wire those servers and someone else will need to expand the AI into more fields.
We humans don't like change much, but that doesn't mean we should pump money to blockbuster.
Quitting is an individual action with potential adverse consequences (misery).
The political question is why despite productivity in economic terms (which i know is flawed) growing many times over, do we still have to work as much, get paid so little, and have so many unemployed people looking for a job?
Looks like without a parasitic capitalist class, we could share resources and work and have people live better lives and work less.
Your job being automated to the point where you no longer proform it has the same adverse consequences. Whether this could work is some economic and societal setup is irrelevant, because that's not the world we live in.
What are you talking about? The only reason we could work less is because we always gave way to innovation. We don't haul large blocks of ice because "evil capitalists" started making fridges. Or do you want to reinstate ice trade, just because it generated more jobs? Or ban tractors because farming by hand was made by way more people?
"Sharing resources" doesn't work. We share way too much just to keep pensioners alive and can't have kids because of that. Confiscating all the billionaire's wealth wouldn't make a dent and would destroy much more.
Not sorry for the rant
> Third, once destroyed those industries will not meaningfully return. This is not a process that will result in a natural ebb and flow. Once the software development profession is disrupted, it will stay that way as talent pipelines dry up and expertise ages out
I can echo that. If/when I am replaced with an LLM, I plan on switching to a hands-on profession - welder, electrician, machinist. I am also not planning on ever going back to software... errr... prompt "engineering". Maybe for twice my pre-LLM salary at a place with a no-LLM policy? More at a prompt-engineering place.
> These tools are here. They’re not going away.
They might very well go away. There is definitely an AI bubble, and it remains to be seen whether it deflates gradually or pops spectacularly. Geoeconomics might destroy them by constraining their access to hardware. The capabilities are real, but whether those capabilities are realised is a different matter.
Deepseek is realistic about this and is offering great prices.
I think that once companies are more realistic about token demand they can start making a profit
What is China subsidises its inference providers to provide like utility level "intelligence" at or below cost globally until all the knowledge jobs in the west go the way of the manufacturing ones..
Even if that were true, the smaller quants that can be run on a MacBook aren't going anywhere
Providing services below cost is how SV and American tech operated last 25 years. It was the same pattern again and again - sell bellow cost, competition cant compete, create quasi monopoly, enshittify and start earning.
OpenAI and Anthropic still sells tokens under cost.
Yup, and the marginal value of the boutique models is much lower than the marginal cost. Most workflows have no need for american models.
I was with the article - especially the bit about the 10% an LLM loses and how that loss can be de-amplified across eventual meaning, but then...
"My first instinct is to laugh and shake my head. One need not look very far to find indignant software developers absolutely certain that their jobs cannot possibly be automated away by the very tools their industry contemporaries are creating to replace them. I suspect you’d also not have to look far into their posting histories to find those same people comparing cabbies to buggy whip makers."
What a rude and callous comment. I'm one of those developers, I'd love to see an LLM even fractionally capable of some of the things my job entails. Laugh at me as I defend my lifelong career, why don't you? I'm also one who decries such things as the removal of local services, taxi and otherwise, for those in the cloud. Screw you, man.
Not the OP but this line really got my attention
> I'd love to see an LLM even fractionally capable of some of the things my job entails
If you don't mind could you please elaborate on what tasks or knowledge your job requires that you feel an LLM isn't even _fractionally_ capable of? I understand if you said that out of emotion and frustration, but if you're serious I'm intensely and genuinely curious because nowadays it seems like the frontier models are capable of more programming tasks than not with the proper harness and engineer. When was the last time you tried?
I just want to emphasize that this is not a further provocation, just potentially in awe of what you do at your job and would like to know more.
classic example of the goomba fallacy - "all software engineers are one single person-caricature I have in my head and are therefore a stupid walking contradiction whenever two of them express different opinions".
Do these hypothetical laid off software engineers with no empathy for buggy drivers exist outside the imagination of people who hate software engineers? I know in my education we had a bunch of mandatory courses about automation and the effects on workers, how to consult with workers on how to lift them up rather than be antagonistic, and consider different stakeholders (where "the business" or "product" was but one of many) etc.
Is this sarcastic or do you just not see the irony?
It's not easy being a developer right now. We went from being dorky and unpopular in school to being superstars (at work) and now the writing's on the wall. I can sympathize, but I'll concede that our recently developed attention for the state of nature and the fair distribution of capital is .. interesting.
I don't disagree, but the absolute hubris to think that our job is particularly difficult is also ridiculous. One by one our responsibilities will be able to be shifted into highly leveragable software, whether it's years or decades away.
There’s no irony. Author made a straw man of some bad, bad developer laughing at poor cabbies and it all clicked once VCs came for the fellow professionals.
Tech bro lamenting and berating developers how immoral and hypocritical they’re for “laughing” (has this ever happened?) at other professions, while doing the same. Story as old as time.
I think you wanted to be offended and didn't even understand the point. Maybe I'm also being rude.
The point of this sentence is to show that there are Software Developers who are looking at this in a short-sighted way. There doesn't seem anything rude about it.
Great piece, well written and succinctly sums up my thoughts.
The bit I still don’t understand is how we all put up with the hallucinations. I was questioning Gemini last night about whether it could analyse a Fourtet song and give me a break down of the structure from beginning to end. “Sure!” it said with the endless enthusiasm you get from Gen tools, and then proceeded to spit out an absolute sack of fabricated shit. I pushed back, it apologised, and then generated more crap that had nothing to do with reality, I pushed back, we looped again, still just total fiction: “the drums don’t come in until bar 16” on a song that opens with a drum loop, that kind of crap.
We’re so so far away from tools here that are anywhere near being trustworthy and accurate. And yet we (including myself) are chunking out code after code. It’s so bizarre.
I’m guessing it’s that humans don’t have capacity to deal with this kind of scenario - it’s like having a junior staff member who is utterly incredible 90% of the time - completely convincing in their certainty and skill level, and then 10% of the time you catch them doing a shit in their desk drawer because they couldn’t be arsed to walk to the toilet. AI’s are basically sociopaths.
Doesn’t the article make the argument that since you can write tests this is not as much of a problem for code gen ?
Its arguable whether it is a foolproof solution (I don’t think so) but it definitely makes it look like you can build a harness around the stochastic machine that will validate the correctness of the generated randomness.
Monkeys and typewriters when you can quickly validate whether it’s Shakespeare or not is a costly but theoretically feasible scenario. No?
> We’re so so far away from tools here that are anywhere near being trustworthy and accurate. And yet we (including myself) are chunking out code after code. It’s so bizarre.
I think one more thing this whole LLM charade in the last few years has revealed is that no-one really cares. As long as it "looks" like it works, turns out, its all fine.
Are people seriously getting use out of LLMs? Everything they produce is extermely sloppy in my experience, like actually mostly useless. I really dont understand the hype. Its very confusing.
I work 40 hours a week at a company I care very little for. Despite that, I still have to make 40 hours. I also have a side-project that I do in my free time. I do not have the energy to sit behind my desk for >40 hrs a week
So I walk, program, sit, get coffee, read a bit, come back and review the code. Most of the time it's fine, sometimes I had forgotten to mention something, and have to correct it, but this step doesn't take more than ~15 mins.
I then have a feature that would've taken me multiple days. Not because I need multiple days, but because I do not have the 8 hours of continuous time to work on it.
People have forgotten that when you start your programming day, you have to get up to speed which takes me longer than others. Let's say this takes ~15 mins. That means that if you spend 2 hours programming, ~12% of that was just getting back into the groove. LLMs do this instantly, it reads context, you can ask it questions, and you're up to speed again in less than a minute.
The point here is not that LLMs provide high quality code, but they do save you a bit of time and energy, which is worth a lot in my opinion.
A lot of inventions haven't been that ground-breaking; only there to save time. You can wash the dishes by hand, but you can also have a machine do it while you go watch TV.
I am convinced the people who swear by them either have very different work tasks from me, or they have very different ideas about what a job well done looks like.
I feel like everything I apply these things to sends me up a much messier and long winded route to a useable result, when compared to just doing it myself from the jump. Even the things they're ostensibly good at like sorting data comes out so messy it's practically net zero by the time you're done with quality control.
I tend to do the main structural parts myself and tell it to fill in the gaps and add tests, which sort of works 70% of the time. It may not be worth what my company is being charged.
That's my approach as well. When I try to mentally compare the cost of the subscription to the performance, I think it's extremely dubious at best.
Every time I have actually engaged in conversations with people making the same claim you do, it turned out they did not really invest time into learning how to work with coding agents: They assumed, given that they’re developers, the know how to code, and thus would knew how to steer a coding agent.
That is a wrong assumption, however. An agent is an entirely new tool in your toolbox, with no similarity to any of those you already have. You will need to learn how to wield it, like a new programming language or technology you’re unfamiliar with. You will need to do some small side projects to learn. You will need to develop a feeling for how it reacts to your inputs, when to reset the session, pass it links to documentation, or interrupt it.
None of this comes intuitively. It takes time and effort, and if you’re not ready to consciously invest that, coding agents are not going to work efficiently for you. That doesn’t take away from their utility though.
No, the more appropriate analogy would be that LLMs (reasoning agents would be better term, as they are no longer just an LLMs) are new beings rather than new tools.
A new tool can be ultimately understood, and has a well-defined behaviour you can rely on. Then you can indeed become an expert in using a new tool.
But AI agent is more a new being (or person). It's not possible to understand them, as they are not meant to. Each of them has individuality, and can change tomorrow. There is no guarantee of a common behaviour.
So you could replace "tool" in your comment with "person" and you will see the flaw of your argument. No matter how many people you saw before, it's difficult to generalize the skillset (at least for humans there are some biological and cultural arguments why they can be generalized over). You can always meet someone different, who's, for whatever reason, not being helpful to you.
Most of the interaction surface with a coding agent is the harness anyway, so I'd definitely classify them as tools. Moreover, I don't like the anthropomorphism of calling them beings—and even if I were to follow that narrative, you can absolutely become better at interacting with people by practicing it.
Regardless, it sure seems like developers experienced with using coding agents achieve better results than those without, which pretty much refuses your entire point.
Yes you can learn the harness, that's a tool. But it's not the important part.
I am not anthroporphizing them. I am just saying people are a better analogy.
> seems like developers experienced with using coding agents achieve better results than those without
That doesn't mean there is a skill involved. The same goes with people - you might just be naturally charismatic and get stuff done better with people. Doesn't mean it's a learnable skill that applies to everyone and every use case.
I don't know, it is pretty simple to type /init in claude and see if its understanding of a project matches reality. And if it can make changes to it.
The difference working with agents, is that a solo dev suddenly becomes a dev manager, who has to deal with a unreliable team of developers far out on the spectrum.
Without you telling what your job entails and how the pristine artifacts of your crystalline mind fit into the larger picture it's very hard to have a conversation about this.
But yes, many developers I know haven't coded manually in months and that includes me. That doesn't mean I drink coffee and take walks while the agent codes however. I'm now in the driving seat instead of wrangling syntax and waiting for my hands to type something.
I manage the rules, the intent, the structure. If I don't like what it does, I update the coding standard, the docs, the specs, you name it. If I want a different architecture I can actually get it done in 10min instead of a week - if at all. It's more effort than coding for me, because coding is slow and methodical and incentivizes pedantry, but for me it's a massive improvement. I have been bored out of my mind for a decade now, but I'm relatively good at what I do so I just stuck with it. I'm one of those "middle management types stuck in developer role"-type of guys so I guess my personality and proclivities have something to do with it. I also know quite a few grumpy old-school types that don't get anything done with LLMs. They can't communicate, they don't understand psychology, they don't understand architecture and have a debilitating case of missing the forest for the trees. The spread, in my experience, is massive.
Just out of curiosity, did you actually use claude?
Otherwise have a look here:
https://news.ycombinator.com/item?id=48406174
Yes. Most of the people I see are on the extremes of either hating LLMs or fanatically loving it, but I am somewhere in the middle ground, I think, where I see it as a tool that sometimes helps to some extent. I don't depend on it, it almost never gives me new ideas or ways to do things, but for trivial tasks it helps me relax supervising it - usually writing the first draft for documentation or suggesting initial updates, even filling in some semi-repetitive code patterns in web apps ("create empty functions and link the buttons to it"). It is just a tool, not too bad to ignore, not too good to get more productivity, sometimes useful. Using it is like using an IDE 30 years ago or Intellisense 15-20 years ago.
LE. I see current versions of LLM like an intern that helps me doing work. We work together, I give directions and supervision and I am responsible for the results. I cannot give complex tasks and I cannot skip checking everything, but it usually helps.
I am getting serious use out of LLMs, and everyone else who does knows that those claiming otherwise are only getting slop because they’re not giving it enough guard rails, likely because they’re uninitiated, likely because they hate AI in the first place.
is a sometimes useful agent, and a sometimes useful slightly smarter search engine.
it can be useful for some automation... but its also dangerously dumb for that.
Meh just another anti AI rant - this time with the cloak of legitimacy taken from a market downturn.
Whether you agree with it or not, the author makes some good points and it goes way beyond just a "rant".