I worry about the "brain atrophy" part, as I've felt this too. And not just atrophy, but even moreso I think it's evolving into "complacency".
Like there have been multiple times now where I wanted the code to look a certain way, but it kept pulling back to the way it wanted to do things. Like if I had stated certain design goals recently it would adhere to them, but after a few iterations it would forget again and go back to its original approach, or mix the two, or whatever. Eventually it was easier just to quit fighting it and let it do things the way it wanted.
What I've seen is that after the initial dopamine rush of being able to do things that would have taken much longer manually, a few iterations of this kind of interaction has slowly led to a disillusionment of the whole project, as AI keeps pushing it in a direction I didn't want.
I think this is especially true if you're trying to experiment with new approaches to things. LLMs are, by definition, biased by what was in their training data. You can shock them out of it momentarily, whish is awesome for a few rounds, but over time the gravitational pull of what's already in their latent space becomes inescapable. (I picture it as working like a giant Sierpinski triangle).
I want to say the end result is very akin to doom scrolling. Doom tabbing? It's like, yeah I could be more creative with just a tad more effort, but the AI is already running and the bar to seeing what the AI will do next is so low, so....
It's not just brain atrophy, I think. I think part of it is that we're actively making a tradeoff to focus on learning how to use the model rather than learning how to use our own brains and work with each other.
This would be fine if not for one thing: the meta-skill of learning to use the LLM depreciates too. Today's LLM is gonna go away someday, the way you have to use it will change. You will be on a forever treadmill, always learning the vagaries of using the new shiny model (and paying for the privilege!)
I'm not going to make myself dependent, let myself atrophy, run on a treadmill forever, for something I happen to rent and can't keep. If I wanted a cheap high that I didn't mind being dependent on, there's more fun ones out there.
> let myself atrophy, run on a treadmill forever, for something
You're lucky to afford the luxury not to atrophy.
It's been almost 4 years since my last software job interview and I know the drills about preparing for one.
Long before LLMs my skills naturally atrophy in my day job.
I remember the good old days of J2ME of writing everything from scratch. Or writing some graph editor for universiry, or some speculative, huffman coding algorithm.
That kept me sharp.
But today I feel like I'm living in that netflix series about people being in Hell and the Devil tricking them they're in Heaven and tormenting them: how on planet Earth do I keep sharp with java, streams, virtual threads, rxjava, tuning the jvm, react, kafka, kafka streams, aws, k8s, helm, jenkins pipelines, CI-CD, ECR, istio issues, in-house service discovery, hierarchical multi-regions, metrics and monitoring, autoscaling, spot instances and multi-arch images, multi-az, reliable and scalable yet as cheap as possible, yet as cloud native as possible, hazelcast and distributed systems, low level postgresql performance tuning, apache iceberg, trino, various in-house frameworks and idioms over all of this?
Oh, and let's not forget the business domain, coding standards, code reviews, mentorships and organazing technical events.
Also, it's 2026 so nobody hires QA or scrum masters anymore so take on those hats as well.
Businesses too. For two years it's been "throw everything into AI." But now that shit is getting real, are they really feeling so coy about letting AI run ahead of their engineering team's ability to manage it? How long will it be until we start seeing outages that just don't get resolved because the engineers have lost the plot?
> It's not just brain atrophy, I think. I think part of it is that we're actively making a tradeoff to focus on learning how to use the model rather than learning how to use our own brains and work with each other.
I agree with the sentiment but I would have framed it differently. The LLM is a tool, just like code completion or a code generator. Right now we focus mainly on how to use a tool, the coding agent, to achieve a goal. This takes place at a strategic level. Prior to the inception of LLMs, we focused mainly on how to write code to achieve a goal. This took place at a tactical level, and required making decisions and paying attention to a multitude of details. With LLMs our focus shifts to a higher-level abstraction. Also, operational concerns change. When writing and maintaining code yourself, you focus on architectures that help you simplify some classes of changes. When using LLMs, your focus shifts to building context and aiding the model effectively implement their changes. The two goals seem related, but are radically different.
I think a fairer description is that with LLMs we stop exercising some skills that are only required or relevant if you are writing your code yourself. It's like driving with an automatic transmission vs manual transmission.
> the meta-skill of learning to use the LLM depreciates too. Today's LLM is gonna go away someday, the way you have to use it will change. You will be on a forever treadmill, always learning the vagaries of using the new shiny model (and paying for the privilege!)
I haven’t found this to be true at all, at least so far.
As models improve I find that I can start dropping old tricks and techniques that were necessary to keep old models in line. Prompts get shorter with each new model improvement.
It’s not really a cycle where you’re re-learning all the time or the information becomes outdated. The same prompt structure techniques are usually portable across LLMs.
I think you have to be aware of how you use any tool but I don’t think this is a forever treadmill. It’s pretty clear to me since early on that the goal is for you the user to not have to craft the perfect prompt. At least for my workflow it’s pretty darn close to that for me.
I have deliberately moderated my use of AI in large part for this reason. For a solid two years now I've been constantly seeing claims of "this model/IDE/Agent/approach/etc is the future of writing code! It makes me 50x more productive, and will do the same for you!" And inevitabely those have all fallen by the wayside and been replaced by some new shiny thing. As someone who doesn't get intrinsic joy out of chasing the latest tech fad I usually move along and wait to see if whatever is being hyped really starts to take over the world.
This isn't to say LLMs won't change software development forever, I think they will. But I doubt anyone has any idea what kind of tools and approaches everyone will be using 5 or 10 years from now, except that I really doubt it will be whatever is being hyped up at this exact moment.
In my experience all technology has been like this though. We are on the treadmill of learning the new thing with our without LLMs. That's what makes tech work so fun and rewarding (for me anyway).
I assume you're living in a city. You're already renting out a lot of things to others (security, electricity, water, food, shelter, transportation), what is different with white collar work?
I think I should write more about but I have been feeling very similar. I've been recently exploring using claude code/codex recently as the "default", so I've decided to implement a side project.
My gripe with AI tools in the past is that the kind of work I do is large and complex and with previous models it just wasn't efficient to either provide enough context or deal with context rot when working on a large application - especially when that application doesn't have a million examples online.
I've been trying to implement a multiplayer game with server authoritative networking in Rust with Bevy. I specifically chose Bevy as the latest version was after Claude's cut off, it had a number of breaking changes, and there aren't a lot of deep examples online.
Overall it's going well, but one downside is that I don't really understand the code "in my bones". If you told me tomorrow that I had optimize latency or if there was a 1 in 100 edge case, not only would I not know where to look, I don't think I could tell you how the game engine works.
In the past, I could not have ever gotten this far without really understanding my tools. Today, I have a semi functional game and, truth be told, I don't even know what an ECS is and what advantages it provides. I really consider this a huge problem: if I had to maintain this in production, if there was a SEV0 bug, am I confident enough I could fix it? Or am I confident the model could figure it out? Or is the model good enough that it could scan the entire code base and intuit a solution? One of these three questions have to be answered or else brain atrophy is a real risk.
I'm worried about that too. If the error is reproducible, the model can eventually figure it out from experience. But a ghost bug that I can't pattern? The model ends up in a "you're absolutely right" loop as it incorrectly guesses different solutions.
> I've been trying to implement a multiplayer game with server authoritative networking in Rust with Bevy. I specifically chose Bevy as the latest version was after Claude's cut off, it had a number of breaking changes, and there aren't a lot of deep examples online.
I am interested in doing something similar (Bevy. not multiplayer).
I had the thought that you ought be able to provide a cargo doc or rust-analyzer equivalent over MCP? This... must exist?
I'm also curious how you test if the game is, um... fun? Maybe it doesn't apply so much for a multiplayer game, I'm thinking of stuff like the enemy patterns and timings in a soulslike, Zelda, etc.
I did use ChatGPT to get some rendering code for a retro RCT/SimCity-style terrain mesh in Bevy and it basically worked, though several times I had to tell it "yeah uh nothing shows up", at which point is said "of course! the problem is..." and then I learned about mesh winding, fine, okay... felt like I was in over my head and decided to go to a 2D game instead so didn't pursue that further.
I ran into similar issues with context rot on a larger backend project recently. I ended up writing a tool that parses the AST to strip out function bodies and only feeds the relevant signatures and type definitions into the prompt.
It cuts down the input tokens significantly which is nice for the monthly bill, but I found the main benefit is that it actually stops the model from getting distracted by existing implementation details. It feels a bit like overengineering but it makes reasoning about the system architecture much more reliable when you don't have to dump the whole codebase into the context window.
"For this invention will produce forgetfulness in the minds of those who learn to use it, because they will not practice their memory. Their trust in writing, produced by external characters which are no part of themselves, will discourage the use of their own memory within them. You have invented an elixir not of memory, but of reminding; and you offer your pupils the appearance of wisdom, not true wisdom, for they will read many things without instruction and will therefore seem [275b] to know many things, when they are for the most part ignorant and hard to get along with, since they are not wise, but only appear wise."
- Socrates on Writing and Reading, Phaedrus 370 BC
If one reads the dialogue, Socrates is not the one "saying" this, but he is telling a story of what King Thamus said to the Egyptian god Theuth, who is the inventor of writing. He is asking the king to give out the writing, but the king is unsure about it.
Its what is known as one of the Socratic "myths," and really just contributes to a web of concepts that leads the dialogue to its ultimate terminus of aporia (being a relatively early Plato dialogue). Socrates, characteristically, doesn't really give his take on writing. In the text, he is just trying to help his friend write a horny love letter/speech!
I can't bring it up right now, but the end of the dialogue has a rather beautiful characterization of writing in the positive, saying that perhaps logos can grow out of writing, like a garden.
I think if pressed Socrates/Plato would say that LLM's are merely doxa machines, incapable of logos. But I am just spitballing.
Presenting this quote without additional commentary is an interesting Rorschach test.
Thankfully more and more people are seriously considering the effects of technology on true wisdom and getting of the "all technological progress clearly is great, look at all these silly unenlightened naysayers from the past" train.
That is interesting because your mental abilities seem to be correlated with orchestrating a bunch of abstractions you have previously mastered. Are these tools making us stupid because we no longer need to master any of these things? Or are they making us smarter because the abstraction is just trusting AI to handle it for us?
It's unclear if you've presented this quote in order to support or criticize the idea that new technologies make us dumber. (Perhaps that's intentional; if so, bravo).
To me, this feels like support. I was never an adult who could not read or write, so I can't check my experience against Socrates' specific concern. But speaking to the idea of memory, I now "outsource" a lot of my memory to my smartphone.
In the past, I would just remember my shopping list, and go to the grocery store and get what I needed. Sure, sometimes I'd forget a thing or two, but it was almost always something unimportant, and rarely was a problem. Now I have my list on my phone, but on many occasions where I don't make a shopping list on my phone, when I get to the grocery store I have a lot of trouble remembering what to get, and sometimes finish shopping, check out, and leave the store, only to suddenly remember something important, and have to go back in.
I don't remember phone numbers anymore. In college (~2000) I had the campus numbers (we didn't have cell phones yet) of at least two dozen friends memorized. Today I know my phone number, my wife's, and my sister's, and that's it. (But I still remember the phone number for the first house I lived in, and we moved out of that house when I was five years old. Interestingly, I don't remember the area code, but I suppose that makes sense, as area codes weren't required for local dialing in the US back in the 80s.)
Now, some of this I will probably ascribe to age: I expect our memory gets more fallible as we get older (I'm in my mid 40s). I used to have all my credit/debit card numbers, and their expiration dates and security codes, memorized (five or six of them), but nowadays I can only manage to remember two of them. (And I usually forget or mix up the expiration dates; fortunately many payment forms don't seem to check, or are lax about it.) But maybe that is due to new technology to some extent: most/all sites where I spend money frequently remember my card for me (and at most only require me to enter the security code). And many also take Paypal or Google Pay, which saves me from having to recall the numbers.
So I think new technology making us "dumber" is a very real thing. I'm not sure if it's a good thing or a bad thing. You could say that, in all of my examples, technology serving the place of memory has freed up mental cycles to remember more important things, so it's a net positive. But I'm not so sure.
My personal counterpoint is Norman's thesis in Things That Make Us Smart.
I've long tried, and mostly failed, to consider the tradeoffs, to be ever mindful that technologies are never neutral (winners & losers), per Postman's Technopoly.
Writing/reading and AI are so categorically different that the only way you could compare them is if you fundamentally misunderstand how both of them work.
And "other people in the past predicted doom about something like this and it didn't happen" is a fallacious non-argument even when the things are comparable.
I've been thinking along these lines. LLMs seem to have arrived right when we were all getting addicted to reels/tic tocks/whatever. For some reason we love to swipe, swipe, swipe, until we get something funny/interesting/shocking, that gives us a short-lasting dopamine hit (or whatever chemicals it is) that feels good for about 1 second, and we want MORE, so we keep swiping.
Using an LLM is almost exactly the same. You get the occasional, "wow! I've never seen it do that before!" moments (whether that thing it just did was even useful or not), get a short hit of feel goods, and then we keep using it trying to get another hit. It keeps providing them at just the right intervals for people to keep them going just like they do with tick tock
I ran into a new problem today: "reading atrophy".
As in if the LLM doesn't know about it, some devs are basically giving up and not even going to RTFM. I literally had to explain to someone today how something works by...reading through the docs and linking them the docs with screenshots and highlighted paragraphs of text.
Still got push back along the lines of "not sure if this will work". It's. Literally. In. The. Docs.
That's not really a new thing now, it just shows differently.
15 years ago I was working in an environment where they had lots of Indians as cheap labour - and the same thing will show up in any environment where you go for hiring a mass of cheap people while looking more at the cost than at qualifications: You pretty much need to trick them into reading stuff that are relevant.
I remember one case where one had a problem they couldn't solve, and couldn't give me enough info to help remotely. In the end I was sitting next to them, and made them read anything showing up on the screen out loud. Took a few tries where they were just closing dialog boxes without reading it, but eventually we had that under control enough that they were able to read the error messages to me, and then went "Oh, so _that's_ the problem?!"
Overall interacting with a LLM feels a lot like interacting with one of them back then, even down to the same excuses ("I didn't break anything in that commit, that test case was never passing") - and my expectation for what I can get out of it is pretty much the same as back then, and approach to interacting with it is pretty similar. It's pretty much an even cheaper unskilled developer, you just need to treat it as such. And you don't pair it up with other unskilled developers.
The mere existence of the phrase "RTFM" shows that this phenomenon was already a thing. LLMs are the worst thing to happen to people who couldn't read before. When HR type people ask what my "superpower" is I'm so tempted to say "I can read", because I honestly feel like it's the only difference between me and people who suck at working independently.
As someone working in technical support for a long time, this has always been the case.
You can have as many extremely detailed and easy to parse gudies, references, etc. there will always be a portion of customers who refuse to read them.
Never could figure out why because they aren't stupid or anything.
> Eventually it was easier just to quit fighting it and let it do things the way it wanted.
I wouldn't have believed it a few tears ago if you told me the industry would one day, in lockstep, decide that shipping more tech-debt is awesome. If the unstated bet doesn't pay off, that is, AI development will outpace the rate it generates cruft, then there will be hell to pay.
Don't worry. This will create the demand for even more powerful models that are able to untangle the mess created by previous models.
Once we realize the kind of mess _those_ models created, well, we'll need even more capable models.
It's a variation on the theme of Kernighan insight about the more "clever" you are while coding the harder it will be to debug.
EDIT: Simplicity is a way out but it's hard under normal circumstances, now with this kind of pressure to ship fast because the colleague with the AI chimp can outperform you, aiming at simplicity will require some widespread understanding
As someone who's been commissioned many times before to work on or salvage "rescue projects" with huge amounts of tech debt, I welcome that day. Still not there yet though I am starting to feel the vibes shifting.
This isn't anything new of course. Previously it was with projects built by looking for the cheapest bidder and letting them loose on an ill-defined problem. And you can just imagine what kind of code that produced. Except the scale is much larger.
My favorite example of this was a project that simply stopped working due to the amount of bugs generated from layers upon layers of bad code that was never addressed. That took around 2 years of work to undo. Roughly 6 months to un-break all the functionality and 6 more months to clean up the core and then start building on top.
The industry decided that decades ago. We may like to talk about quality and forethought, but when you actually go to work, you quickly discover it doesn't matter. Small companies tell you "we gotta go fast", large companies demand clear OKRs and focusing on actually delivering impact - either way, no one cares about tech debt, because they see it as unavoidable fact of life. Even more so now, as ZIRP went away and no one can afford to pay devs to polish the turd ad infinitum. The mantra is, ship it and do the next thing, clean up the old thing if it ever becomes a problem.
And guess what, I'm finally convinced they're right.
Consider: it's been that way for decades. We may tell ourselves good developers write quality code given the chance, but the truth is, the median programmer is a junior with <5 years of experience, and they cannot write quality code to save their life. That's purely the consequence of rapid growth of software industry itself. ~all production code in the past few decades was written by juniors, it continues to be so today; those who advance to senior level end up mostly tutoring new juniors instead of coding.
Or, all that put another way: tech debt is not wrong. It's a tool, a trade-off. It's perfectly fine to be loaded with it, if taking it lets you move forward and earn enough to afford paying installments when they're due. Like with housing: you're better off buying it with lump payment, or off savings in treasury bonds, but few have that money on hand and life is finite, so people just get a mortgage and move on.
--
Edited to add: There's a silver lining, though. LLMs make tech debt legible and quantifiable.
LLMs are affected by tech debt even more than human devs are, because (currently) they're dumber, they have less cognitive capability around abstractions and generalizations[0]. They make up for it by working much faster - which is a curse in terms of amplifying tech debt, but also a blessing, because you can literally see them slowing down.
Developer productivity is hard to measure in large part because the process is invisible (happens in people's heads and notes), and cause-and-effect chains play out over weeks or months. LLM agents compress that to hours to days, and the process itself is laid bare in the chat transcript, easy to inspect and analyze.
The way I see it, LLMs will finally allow us to turn software development at tactical level from art into an engineering process. Though it might be too late for it to be of any use to human devs.
--
[0] - At least the out-of-distribution ones - quirks unique to particular codebase and people behind it.
> I wouldn't have believed it a few tears ago if you told me the industry would one day, in lockstep, decide that shipping more tech-debt is awesome.
It's not debt if you never have to pay it back. If a model can regenerate a whole relibale codebase in minutes from a spec, then your assessment of "tech debt" in that output becomes meaningless.
My disillusionment comes from the feeling I am just cosplaying my job. There is nothing to distinguish one cosplayer from another. I am just doordashing software, at this point, and I'm not in control.
I don’t get this at all. I’m using LLM’s all day and I’m constantly having to make smart architectural choices that other less experienced devs won’t be making. Are you just prompting and going with whatever the initial output is, letting the LLM make decisions? Every moderately sized task should start with a plan, I can spend hours planning, going off and thinking, coming back to the plan and adding/changing things, etc. Sometimes it will be days before I tell the LLM to “go”. I’m also constantly optimising the context available to the LLM, and making more specific skills to improve results. It’s very clear to me that knowledge and effort is still crucial to good long term output… Not everyone will get the same results, in fact everyone is NOT getting the same results, you can see this by reading the wildly different feedback on HN. To some LLM’s are a force multiplier while others claim they can’t get a single piece of decent output…
I think the way you’re using these tools that makes you feel this way is a choice. You’re choosing to not be in control and do as little as possible.
100% there....it's getting to a point where a project manager reports a bug AND also pastes a response from Claude (he ran Claude against our codebase) on how to fix the bug..Like I'm just copying what Claude said and making sure the thing compiles (.NET). What makes me sleep at night...for now is the fact that Claude isn't supporting 9pm deployments and AWS Infra support ...it's already writing code but not supporting it yet...
What kind of software are you writing? Are you just a "code monkey" implementing perfectly described Jira tickets (no offense meant)? I cannot imagine feeling this way with what I'm working on, writing code is just a small part of it, most of the time is spent trying to figure out how to integrate the various (undocumented and actively evolving) external services involved together in a coherent, maintainable and resilient way. LLMs absolutely cannot figure this out themselves, I have to figure it out myself and then write it all in its context, and even then it mostly comes up with sub-par, unmaintainable solutions if I wasn't being precise engouh.
They are amazing for side projects but not for serious code with real world impact where most of the context is in multiple people's head.
I find the atrophy and zoning out or context switching problematic, because it takes a few seconds/ minutes in "thinking" and then BAM! I have 500 lines of all sorts of buggy and problematic code to review and get a sycophantic, not-enough-mature entity to correct.
At some point, I find myself needing to disconnect out of overwhelm and frustration. Faster responses isn't necessarily better. I want more observability in the development process so that I can be a party to it. I really have felt that I need to orchestrate multiple agents working in tandem, playing sort of a bad-cop, good-cop and a maybe a third trying to moderate that discussion and get a fourth to effectively incorporate a human in the mix. But that's too much to integrate in my day job.
I’ve actually found the tool that inspires the most worry about brain atrophy to be Copilot. Vscode is full of flashing suggestions all over. A couple days ago, I wanted to write a very quick program, and it was basically impossible to write any of it without Copilot suggesting a whole series of ways to do what it thought I was doing. And it seems that MS wants this: the obvious control to turn it off is actually just “snooze.”
I found the setting and turned it off for real. Good riddance. I’ll use the hotkey on occasion.
My experience is the opposite - I haven't used my brain more in a while.. Typing characters was never what developers were valued for anyway. The joy of building is back too.
> Like there have been multiple times now where I wanted the code to look a certain way, but it kept pulling back to the way it wanted to do things. Like if I had stated certain design goals recently it would adhere to them, but after a few iterations it would forget again and go back to its original approach, or mix the two, or whatever. Eventually it was easier just to quit fighting it and let it do things the way it wanted.
Absolutely. At a certain level of usage, you just have to let it do it's thing.
People are going to take issue with that. You absolutely don't have to let it do its thing. In that case you have to be way more in the loop. Which isn't necessarily a bad thing.
But assuming you want it to basically do everything while you direct it, it becomes pointless to manage certain details. One thing in my experience is that Claude always wants to use ReactRouter. My personal preference is TanStack router, so I asked it to use it initially. That never really created any problems but after like the 3rd time of realizing I forget to specify it, I also realized that it's totally pointless. ReactRouter works fine and Claude uses it fine - its pointless to specify otherwise.
> I want to say it's very akin to doom scrolling. Doom tabbing? It's like, yeah I could be more creative with just a tad more effort, but the AI is already running and the bar to seeing what the AI will do next is so low, so....
Yea exactly, Like we are just waiting so that it gets completed and after it gets completed then what? We ask it to do new things again.
Just as how if we are doom scrolling, we watch something for a minute then scroll down and watch something new again.
The whole notion of progress feels completely fake with this. Somehow I guess I was in a bubble of time where I had always end up using AI in web browsers (just as when chatgpt 3 came) and my workflow didn't change because it was free but recently changed it when some new free services dropped.
"Doom-tabbing" or complete out of the loop AI agentic programming just feels really weird to me sucking the joy & I wouldn't even consider myself a guy particular interested in writing code as I had been using AI to write code for a long time.
I think the problem for me was that I always considered myself a computer tinker before coder. So when AI came for coding, my tinkering skills were given a boost (I could make projects of curiosity I couldn't earlier) but now with AI agents in this autonomous esque way, it has come for my tinkering & I do feel replaced or just feel like my ability of tinkering and my interests and my knowledge and my experience is just not taken up into account if AI agent will write the whole code in multi file structure, run commands and then deploy it straight to a website.
I mean my point is tinkering was an active hobby, now its becoming a passive hobby, doom-tinkering? I feel like I have caught up on the feeling a bit earlier with just vibe from my heart but is it just me who feels this or?
Another thing I’ve experienced is scope creep into the average. Both Claude and ChatGPT keep making recommendations and suggestions that turns the original request into something that resembles other typical features. Sometimes that’s a good thing, because it means I’ve missed something. A lot of times, especially when I’m just riffing on ideas, it turns into something mundane and ordinary and I’ll have lost my earlier train of thought.
A quick example is trying to build a simple expenses app with it. I just want to store a list of transactions with it. I’ve already written the types and data model and just need the AI to give me the plumping. And it will always end up inserting recommendations about double entry bookkeeping.
yeah but that's like recommending a webserver for your Internet facing website. If you want to give an example of scope creep, you need a better example than double entry book keeping for an accounting app.
> I worry about the "brain atrophy" part, as I've felt this too. And not just atrophy, but even moreso I think it's evolving into "complacency".
Not trusting the ML's output is step one here, that keeps you intellectually involved - but it's still a far cry from solving the majority of problems yourself (instead you only solve problems ML did a poor job at).
Step two: I delineate interesting and uninteresting work, and Claude becomes a pair programmer without keyboard access for the latter - I bounce ideas off of it etc. making it an intelligent rubber duck. [Edit to clarify, a caveat is that] I do not bore myself with trivialities such as retrieving a customer from the DB in a REST call (but again, I do verify the output).
> I do not bore myself with trivialities such as retrieving a customer from the DB in a REST call
Genuine question, why isn't your ORM doing that? I see a lot of use cases for LLMs that seem to be more expensive ways to do snippets and frameworks...
I've gone years without coding and when I come back to it, it's like riding a bike! In each iteration of my coding career, I have become a better developer, even after a large gap. Now I can "code" during my gap. Were I ever to hand-code again, I'm sure my skills would be there. They don't atrophy, like your ability to ride a bike doesn't atrophy. Yes you may need to warm back up, but all the connections in your brain are still there.
Have you ever learnt a foreign language (say Mongolian, or Danish) and then never spoken it, nor even read anything in it for over 10 years? It is not like riding a bike, it doesn’t just come back like that. You have to actually relearn the language, practice it, and you will suck at it for months. Comprehension comes first (within weeks) but you will be speaking with grammatical errors, mispronunciations, etc. for much longer. You won‘t have to learn the language from scratch, second time around is much easier, but you will have to put in the effort. And if you use google translate instead of your brain, you won‘t relearn the language at all. You will simply forget it.
You might still have the skillset to write code, but depending on length of the break your knowledge of tools, frameworks, patterns would be fairly outdated.
I used to know a person like that - high in the company structure who would claim he was a great engineer, but all the actual engineers would make jokes about him and his ancient skills during private conversations.
I feel like I'm still a couple steps behind in skill level as my lead and is trying to gain more experience I do wonder if I am shooting myself in the foot if I rely too much on AI at this stage. The senior engineer I'm trying to learn from can very effectively use ai because he has very good judgement of code quality, I feel like if I use AI too much I might lose out on chance to improve my judgement. It's a hard dilemma.
Honestly, this seems very much like the jump from being an individual contributor to being an engineering manager.
The time it happened for me was rather abrupt, with no training in between, and the feeling was eerily similar.
You know _exactly_ why the best solution is, you talk to your reports, but they have minds of their own, as well as egos, and they do things … their own way.
At some point I stopped obsessing with details and was just giving guidance and direction only in the cases where it really mattered, or when asked, but let people make their own mistakes.
Now LLMs don’t really learn on their own or anything, but the feeling of “letting go of small trivial things” is sorta similar. You concentrate on the bigger picture, and if it chose to do an iterative for loop instead of using a functional approach the way you like it … well the tests still pass, don’t they.
The only issue is that as an engineering manager you reasonably expect that the team learns new things, improve their skills, in general grow as engineers. With AI and its context handling you're working with a team where each member has severe brain damage that affects their ability to form long term memories.
You can rewire their brain to a degree teaching them new "skills" or giving them new tools, but they still don't actually learn from their mistakes or their experiences.
You could probably combat this somewhat with a skill that references to examples of the code you don't want and the code you do. And then each time you tell it to correct the code you ask it to put that example into the references.
You then tell your agent to always run that skill prior to moving on. If the examples are pattern matchable you can even have the agent write custom lints if your linter supports extension or even write a poor man’s linter using ast-grep.
I usually have a second session running that is mainly there to audit the code and help me add and adjust skills while I keep the main session on the task of working on the feature. I've found this far easier to stay engaged than context switching between unrelated tasks.
> Like if I had stated certain design goals recently it would adhere to them, but after a few iterations it would forget again and go back to its original approach, or mix the two, or whatever.
Context management, proper prompting and clear instructions, proper documentation are still relevant.
The solution for brain atrophy I personally arrived is to use coding agents at work, where, let’s be honest, velocity is a top priority and code purity doesn’t matter that much. Since we use stack I super familial with, I can quite fast verify produced code and tweak it if needed.
However, for hobby projects where I purposely use tech I’m not very familiar with, I force myself not to use LLMs at all - even as a chat. Thus, operating The old way - writing code manually, reading documentation, etc brings me a joy of learning back and, hopefully, establishes new neurone connections.
I think this is where tools like OpenSpec [1] may help. The deterioration in quality is because the context is degrading, often due to incomplete or amibiguous requests from the coder. With a more disciplined way of creating and persisting locally the specs for the work, especially if the agent got involved in creating that too, you'll have a much better chance of keeping the agent focussed and aligned.
> AI keeps pushing it in a direction I didn't want
The AI definitely has preferences and attention issues, but there are ways to overcome them.
Defining code styles in a design doc, and setting up initial examples in key files goes a long way. Claude seems pretty happy to follow existing patterns under these conditions unless context is strained.
I have pretty good results using a structured workflow that runs a core loop of steps on each change, with a hook that injects instructions to keep attention focused.
In the happy case where I have a good idea of the changes necessary, I will ask it to do small things, step by step, and examine what it does and commit.
In the unhappy case where one is faced with a massive codebase and no idea where to start, I find asking it to just “do the thing” generates slop, but enough for me to use as inspiration for the above.
yeah, because the thing is: at the end of the day: laying things out the way LLMs can understand is becoming more important than doing them the “right” way— a more insidious form of the same complacency. and one in which i am absolutely complicit.
> LLM coding will split up engineers based on those who primarily liked coding and those who primarily liked building.
I’ve always said I’m a builder even though I’ve also enjoyed programming (but for an outcome, never for the sake of the code)
This perfectly sums up what I’ve been observing between people like me (builders) who are ecstatic about this new world and programmers who talk about the craft of programming, sometimes butting heads.
One viewpoint isn’t necessarily more valid, just a difference of wiring.
I noticed the same thing, but wasn't able to put it into words before reading that. Been experimenting with LLM-based coding just so I can understand it and talk intelligently about it (instead of just being that grouchy curmudgeon), and the thought in the back of my mind while using Claude Code is always:
"I got into programming because I like programming, not whatever this is..."
Yes, I'm building stupid things faster, but I didn't get into programming because I wanted to build tons of things. I got into it for the thrill of defining a problem in terms of data structures and instructions a computer could understand, entering those instructions into the computer, and then watching victoriously while those instructions were executed.
If I was intellectually excited about telling something to do this for me, I'd have gotten into management.
Same same. Writing the actual code is always a huge motivator behind my side projects. Yes, producing the outcome is important, but the journey taken to get there is a lot of fun for me.
I used Claude Code to implement a OpenAI 4o-vision powered receipt scanning feature in an expense tracking tool I wrote by hand four years ago. It did it in two or three shots while taking my codebase into account.
It was very neat, and it works great [^0], but I can't latch onto the idea of writing code this way. Powering through bugs while implementing a new library or learning how to optimize my test suite in a new language is thrilling.
Unfortunately (for me), it's not hard at all to see how the "builders" that see code as a means to an end would LOVE this, and businesses want builders, not crafters.
In effect, knowing the fundamentals is getting devalued at a rate I've never seen before.
[^0] Before I used Claude to implement this feature, my workflow for processing receipts looked like this: Tap iOS Shortcut, enter the amount, snap a pic of the receipt, type up the merchant, amount and description for the expense, then have the shortcut POST that to my expenses tracking toolkit which, then, POSTs that into a Google Sheet. This feature amounted the need for me to enter the merchant and amount. Unfortunately, it often took more time to confirm that the merchant, amount and date details OpenAI provided were correct (and correct it when details were wrong, which was most of the the time) than it did to type out those details manually, so I just went back to my manual workflow. However, the temptation to just glance at the details and tap "This looks correct" was extremely high, even if the info it generated was completely wrong! It's the perfect analogue to what I've been witnessing throughout the rise of the LLMs.
Same. This kind of coding feels like it got rid of the building aspect of programming that always felt nice, and it replaced it entirely with business logic concerns, product requirements, code reviews, etc. All the stuff I can generally take or leave. It's like I'm always in a meeting.
>If I was intellectually excited about telling something to do this for me, I'd have gotten into management.
Exactly this. This is the simplest and tersest way of explaining it yet.
What I have enjoyed about programming is being able to get the computer to do exactly what I want. The possibilities are bounded by only what I can conceive in my mind. I feel like with AI that can happen faster.
This gets at the heart of the quality of results issues a lot of people are talking about elsewhere here. Right now, if you treat them as a system where you can tell it what you want and it will do it for you, you're building a sandcastle. Instead of that, also describe the correct data structures and appropriate algorithms to use against them, as well as the particulars of how you want the problem solved, it's a different situation altogether. Like most systems, the quality of output is in some way determined by the quality of input.
There is a strange insistence on not helping the LLM arrive at the best outcome in the subtext to this question a lot of times. I feel like we are living through the John Henry legend in real time
> I got into it for the thrill of defining a problem in terms of data structures and instructions a computer could understand, entering those instructions into the computer, and then watching victoriously while those instructions were executed.
You can still do that with Claude Code. In fact, Claude Code works best the more granular your instructions get.
IMO, this isn't entirely a "new world" either, it is just a new domain where the conversation amplifies the opinions even more (weird how that is happening in a lot of places)
What I mean by that: you had compiled vs interpreted languages, you had types vs untyped, testing strategies, all that, at least in some part, was a conversation about the tradeoffs between moving fast/shipping and maintainability.
But it isn't just tech, it is also in methodologies and the words use, from "build fast and break things" and "yagni" to "design patterns" and "abstractions"
As you say, it is a different viewpoint... but my biggest concern with where are as industry is that these are not just "equally valid" viewpoints of how to build software... it is quite literally different stages of software, that, AFAICT, pretty much all successful software has to go through.
Much of my career has been spent in teams at companies with products that are undergoing the transition from "hip app built by scrappy team" to "profitable, reliable software" and it is painful. Going from something where you have 5 people who know all the ins and outs and can fix serious bugs or ship features in a few days to something that has easy clean boundaries to scale to 100 engineers of a wide range of familiarities with the tech, the problem domain, skill levels, and opinions is just really hard. I am not convinced yet that AI will solve the problem, and I am also unsure it doesn't risk making it worse (at least in the short term)
Much of my career has been spent in teams at companies with products that are undergoing the transition from "hip app built by scrappy team" to "profitable, reliable software" and it is painful. Going from something where you have 5 people who know all the ins and outs and can fix serious bugs or ship features in a few days to something that has easy clean boundaries to scale to 100 engineers of a wide range of familiarities with the tech, the problem domain, skill levels, and opinions is just really hard. I am not convinced yet that AI will solve the problem, and I am also unsure it doesn't risk making it worse (at least in the short term)
“””
This perspective is crucial. Scale is the great equalizer / demoralizer, scale of the org and scale of the systems. Systems become complex quickly, and verifiability of correctness and function becomes harder. Companies that built from day with AI and have AI influencing them as they scale, where does complexity begin to run up against the limitations of AI and cause regression? Or if all goes well, amplification?
But how can you be a responsible builder if you don't have trust in the LLMs doing the "right thing"? Suppose you're the head of a software team where you've picked up the best candidates for a given project, in that scenario I can see how one is able to trust the team members to orchestrate the implementation of your ideas and intentions, with you not being intimately familiar with the details.
Can we place the same trust in LLM agents? I'm not sure. Even if one could somehow prove that LLM are very reliable, the fact an AI agents aren't accountable beings renders the whole situation vastly different than the human equivalent.
I test all of the code I produce via LLMs, usually doing fairly tight cycles. I also review the unit test coverage manually, so that I have a decent sense that it really is testing things - the goal is less perfect unit tests and more just quickly catching regressions. If I have a lot of complex workflows that need testing, I'll have it write unit tests and spell out the specific edge cases I'm worried about, or setup cheat codes I can invoke to test those workflows out in the UI/CLI.
Trust comes from using them often - you get a feeling for what a model is good and bad at, and what LLMs in general are good and bad at. Most of them are a bit of a mess when it comes to UI design, for instance, but they can throw together a perfectly serviceable "About This" HTML page. Any long-form text they write (such as that About page) is probably trash, but that's super-easy to edit manually. You can often just edit down what they write: they're actually decent writers, just very verbose and unfocused.
I find it similar to management: you have to learn how each employee works. Unless you're in the Top 1%, you can't rely on every employee giving 110% and always producing perfect PRs. Bugs happen, and even NASA-strictness doesn't bring that down to zero.
And just like management, some models are going to be the wrong employee for you because they think your style guide is stupid and keep writing code how they think it should be written.
You don't simply put a body in a seat and get software. There are entire systems enabling this trust: college, resume, samples, referral, interviews, tests and CI, monitoring, mentoring, and performance feedback.
And accountability can still exist? Is the engineer that created or reviewed a Pull Request using Claude Code less accountable then one that used PICO?
I think he's really getting at something there. I've been thinking about this a lot (in the context of trying to understand the persistent-on-HN skepticism about LLMs), and the framing I came up with[1] is top-down vs. bottom-up dev styles, aka architecting code and then filling in implementations, vs. writing code and having architecture evolve.
I remember leaving university going into my first engineering job, thinking "Where is all the engineering? All the problem solving and building complex system? All the math and science? Have I been demoted to a lowly programmer?"
Took me a few years to realize that this wasn't a universal feeling, and that many others found the programming tasks more fulfilling than any challenging engineering. I suppose this is merely another manifestation of the same phenomena.
Maybe there's an intermediate category: people who like designing software? I personally find system design more engaging than coding (even though I enjoy coding as well). That's different from just producing an opaque artifact that seems to solve my problem.
So far I haven't seen it actually be effective at "building" in a work context with any complexity, and this despite some on our team desperately trying to make that the case.
I have! You have to be realistic about the projects. The more irreducible local context it needs, the less useful it will be. Great for greenfield code, oneshots, write once read once run for months.
Agreed. I don’t care for engineering or coding, and would gladly give it up the moment I can. I’m also running a one man business where every hour counts (and where I’m responsible for maintaining every feature).
The fact of the matter is LLMs produce lower quality at higher volumes in more time than it would take to write it myself, and I’m a very mediocre engineer.
I find this seperation of “coding” vs “building” so offensive. It’s basically just saying some people are only concerned with “inputs”, while others with “outputs”. This kind of rhetoric is so toxic.
It’s like saying LLM art is separating people into people who like to scribble, and people who like to make art.
I think the division is more likely tied to writing. You have to fundamentally change how you do your job, from one of writing a formal language for a compiler to one of writing natural language for a junior-goldfish-memory-allstar-developer, closer to management then to contributor.
This distinction to me separates the two primary camps
> > LLM coding will split up engineers based on those who primarily liked coding and those who primarily liked building.
> I’ve always said I’m a builder even though I’ve also enjoyed programming (but for an outcome, never for the sake of the code)
> This perfectly sums up what I’ve been observing between people like me (builders) who are ecstatic about this new world and programmers who talk about the craft of programming, sometimes butting heads.
That's one take, sure, but it's a specially crafted one to make you feel good about your position in this argument.
The counter-argument is that LLM coding splits up engineers based on those who primarily like engineering and those who like managing.
You're obviously one of the latter. I, OTOH, prefer engineering.
I prefer engineering too, I tried management and I hated it.
It's just the level of engineering we're split on. I like the type of engineering where I figure out the flow of data, maybe the data structures and how they move through the system.
Writing the code to do that is the most boring part of my job. The LLM does it now. I _know_ how to do it, I just don't want to.
It all boils down to communication in a way. Can you communicate what you want in a way others (in this case a language model) understands? And the parts you can't communicate in a human language, can you use tools to define those (linters, formatters, editorconfig)?
I've done all that with actual humans for ... a decade? So applying the exact same thing to a machine is weirdly more efficient, it doesn't complain about the way I like to have my curly braces - it just copies the defined style. With humans I've found out that using impersonal tooling to inspect code style and flaws has a lot less friction than complaining about it in PR reviews. If the CI computer says no, people don't complain, they fix it.
I feel like this is the core issue that will actually stall LLM coding tools short of actually replacing coding work at large.
'Coders' make 'builders' keep the source code good enough so that 'builders' can continue building without breaking what they built.
If 'builders' become x10 productive and 'coders' become unable to keep up with unsurmountable pile of unmaintainable mess that 'builders' proudly churn out, 'bullders' will start to run into impossibility to build further without starting over and over again hoping that agents will be able to get it right this time.
Yeah, I think this is a bit of insight I had not realized / been able to word correctly yet. There's developers who can let Claud go at it, and be fearless about it like me (though I mostly do it for side projects, but WOW) and then there's developers who will use it like a hammer or axe to help cut down or mold whatever is in their path.
I think both approaches are okay, the biggest thing for me is the former needs to test way more, and review the code more, as developers we don't read code enough, with the "prompt and forget" approach we have a lot of free time we could spend reading the code, asking the model to refactor and refine the code. I am shocked when I hear about hundreds of thousands of lines in some projects. I've rebuilt Beads from the ground up and I'm under 10 lines of code.
So we're going to have various level of AI Code Builders if you will: Junior, Mid, Senior, Architect. I don't know if models will ever pick up the slack for Juniors any time soon. We would need massive context windows for models, and who will pay for that? We need a major AI breakthrough to where the cost goes down drastically before that becomes profitable.
We have services deployed globally serving millions of customers where rigor is really important.
And we have internal users who're building browser extensions with AI that provide valuable information about the interface they're looking at including links to the internal record management, and key metadata that's affecting content placement.
These tools could be handed out on Zip drives in the street and it would just show our users some of the metadata already being served up to them, but it's amazing to strip out 75% of the process of certain things and just have our user (in this case though, it's one user who is driving all of this, so it does take some technical inclination) build out these tools that save our editors so much time when doing this before would have been months and months and months of discovery and coordination and designs that probably wouldn't actually be as useful in the end after the wants of the user are diluted through 18 layers of process.
I like building, but I don't fool myself into thinking it can be done by taking shortcuts. You could build something that looks like a house for half the cost but it won't be structurally sound. That's why I care about the details. Someone has to.
The new LLM centered workflow is really just a management job now.
Managers and project managers are valuable roles and have important skill sets. But there's really very little connection with the role of software development that used to exist.
It's a bit odd to me to include both of these roles under a single label of "builders", as they have so little in common.
I enjoy both and have ended up using AI a lot differently than vibe coders. I rarely use it for generating implementations, but I use it extensively for helping me understand docs/apis and more importantly, for debugging. AI saves me so much time trying to figure out why things aren’t working and in code review.
I deliberately avoid full vibe coding since I think doing so will rust my skills as a programmer. It also really doesn’t save much time in my experience. Once I have a design in mind, implementation is not the hard part.
There's more to it than just coding Vs building though.
For a long time in my career now I've been in a situation where I'd be able to build more if I was willing to abstract myself and become a slide-merchant/coalition-builder. I don't want to do this though.
Yet, I'm still quite an enthusiastic vibe-coder.
I think it's less about coding Vs building and more about tolerance for abstraction and politics. And I don't think there are that many people who are so intolerant of abstraction that they won't let agents write a bunch of code for them.
I’ve heard something similar: “there are people who enjoy the process, and people who enjoy the outcome”. I think this saying comes moreso from artistic circles.
I’ve always considered myself a “process” person, I would even get hung-up on certain projects because I enjoyed crafting them so much.
LLM’s have taken a bit of that “process” enjoyment from me, but I think have also forced some more “outcome” thinking into my head, which I’m taking as a positive.
To me this is similar to car enthusiasms. Some people absolutely love to build their project car, it's a major part of the hobby for them. Others just love the experience of driving, so they buy ready cars or just pay someone to work on the car.
agree completely. I used to be (and still would love to be) a process person, enjoying hand writing bulletproof artisanal code. Switching to startups many years ago gave me a whole new perspective, and its been interesting the struggle between writing code and shipping. Especially when you dont know how long the code you are writing will actually live. LLMs are fantastic in that space.
> I enjoy both and have ended up using AI a lot differently than vibe coders. I rarely use it for generating implementations, but I use it extensively for helping me understand docs/apis and more importantly, for debugging. AI saves me so much time trying to figure out why things aren’t working and in code review.
I had felt like this and still do but man, at some point, I feel like the management churn feels real & I just feel suffering from a new problem.
Suppose, I actually end up having services literally deployed from a single prompt nothing else. Earlier I used to have AI write code but I was interested in the deployment and everything around it, now there are services which do that really neatly for you (I also really didn't give into the agent hype and mostly used browsers LLM)
Like on one hand you feel more free to build projects but the whole joy of project completely got reduced.
I mean, I guess I am one of the junior dev's so to me AI writing code on topics I didn't know/prototyping felt awesome.
I mean I was still involved in say copy pasting or looking at the code it generates. Seeing the errors and sometimes trying things out myself. If AI is doing all that too, idk
For some reason, recently I have been disinterested in AI. I have used it quite a lot for prototyping but I feel like this complete out of the loop programming just very off to me with recent services.
I also feel like there is this sense of if I buy for some AI thing, to maximally extract "value" out of it.
I guess the issue could be that I can have vague terms or have a very small text file as input (like just do X alternative in Y lang) and I am now unable to understand the architectural decisions and the overwhelmed-ness out of it.
Probably gonna take either spec-driven development where I clearly define the architecture or development where I saw something primagen do recently which is that the AI will only manipulate code of that particular function, (I am imagining it for a file as well) and somehow I feel like its something that I could enjoy more because right now it feels like I don't know what I have built at times.
When I prototype with single file projects using say browser for funsies/any idea. I get some idea of what the code kind of uses with its dependencies and functions names from start/end even if I didn't look at the middle
A bit of ramble I guess but the thing which kind of is making me feel this is that I was talking to somebody and shwocasing them some service where AI + server is there and they asked for something in a prompt and I wrote it. Then I let it do its job but I was also thinking how I would architect it (it was some detect food and then find BMR, and I was thinking first to use any api but then I thought that meh it might be hard, why not use AI vision models, okay what's the best, gemini seems good/cheap)
and I went to the coding thing to see what it did and it actually went even beyond by using the free tier of gemini (which I guess didn't end up working could be some rate limit of my own key but honestly it would've been the thing I would've tried too)
So like, I used to pride myself on the architectural decisions I make even if AI could write code faster but now that is taken away as well.
I really don't want to read AI code so much so honestly at this point, I might as well write code myself and learn hands on but I have a problem with build fast in public like attitude that I have & just not finding it fun.
I feel like I should do a more active job in my projects & I am really just figuring out what's the perfect way to use AI in such contexts & when to use how much.
I retired from paid sw dev work in 2020 when COVID arrived.
I’ve worked on my small projects since with all development by hand. I’d followed the rise of AI, but not used it.
Late last year I started a project that included reverse engineering some firmware that runs on an Intel 8096 based embedded processor. I’d never worked on that processor before. There are tools available, but they cost many $. So, I started to think about a simple disassembler.
2 weeks ago we decided to try Claude to see what it could do. We now have a disassembler, assembler and a partially working emulator. No doubt there are bugs and missing features and the code is a bit messy, but boy has it sped up the work.
One thing did occur to me. Vendors of small utilities could be in trouble. For example I needed to cut out some pages from a pdf. I could have found a tool online(I’m sure there are several), write one myself. However, Claude quickly performed the task.
This is a mix of the “in the future, everyone will have a 3D printer at home and just 3D print random parts they need” and “anyone can trivially build Dropbox with rsync themselves” arguments.
Tech savvy users who know how to use LLMs aren’t how vendors of small utilities stay in business.
They stay in business because they sell things to users who are truly clueless with tech (99% of the population, which can’t even figure out the settings app on their phone), and solid distribution/marketing is how you reach those users and can’t really be trivially hacked because everyone is trying to hack it.
Or they stay in business because they offer some sort of guarantee (whether legal, technical, or other) that the users don’t want to burden themselves with because they have other, more important stuff to worry about.
I don't know. It's one thing to tell Joe or Jane User to "Get an FTP account, mount it locally with curlftpfs, and then use SVN or CVS on the mounted filesystem." But if Joe or Jane can just cut-and-paste that advice into a prompt and get their own personal Dropbox...
> Vendors of small utilities could be in trouble. For example I needed to cut out some pages from a pdf. I could have found a tool online(I’m sure there are several), write one myself. However, Claude quickly performed the task.
Definitely. Making small, single-purpose utilities with LLMs is almost as easy these days as googling for them on-line - much easier, in fact, if you account for time spent filtering out all the malware, adware, "to finish the process, register an account" and plain broken "tools" that dominate SERP.
Case in point, last time my wife needed to generate a few QR codes for some printouts for an NGO event, I just had LLM make one as a static, single-page client-side tool and hosted it myself -- because that was the fastest way to guarantee it's fast, reliable, free of surveillance economy bullshit, and doesn't employ URL shorteners (surprisingly common pattern that sometimes becomes a nasty problem down the line; see e.g. a high-profile case of some QR codes on food products leading to porn sites after shortlink got recycled).
> You realize that stamina is a core bottleneck to work
There has been a lot of research that shows that grit is far more correlated to success than intelligence. This is an interesting way to show something similar.
AIs have endless grit (or at least as endless as your budget). They may outperform us simply because they don't ever get tired and give up.
Full quote for context:
Tenacity. It's so interesting to watch an agent relentlessly work at something. They never get tired, they never get demoralized, they just keep going and trying things where a person would have given up long ago to fight another day. It's a "feel the AGI" moment to watch it struggle with something for a long time just to come out victorious 30 minutes later. You realize that stamina is a core bottleneck to work and that with LLMs in hand it has been dramatically increased.
>They never get tired, they never get demoralized, they just keep going and trying things where a person would have given up long ago to fight another day.
"Listen, and understand! That Terminator is out there! It can't be bargained with. It can't be reasoned with. It doesn't feel pity, or remorse, or fear. And it absolutely will not stop... ever, until you are dead!"
If I tell it to implement something it will sometimes declare their work done before it's done. But if I give Claude Code a verifiable goal like making the unit tests pass it will work tirelessly until that goal is achieved. I don't always like the solution, but the tenacity everyone is talking about is there
> If you ever work with LLMs you know that they quite frequently give up.
If you try to single shot something perhaps. But with multiple shots, or an agent swarm where one agent tells another to try again, it'll keep going until it has a working solution.
Context matters, for an LLM just like a person. When I wrote code I'd add TODOs because we cannot context switch to another problem we see every time.
But you can keep the agent fixated on the task AND have it create these TODOs, but ultimately it is your responsibility to find them and fix them (with another agent).
Using LLMs to clean those up is part of the workflow that you're responsible for (... for now). If you're hoping to get ideal results in a single inference, forget it.
I realized a long time ago that I’m better at computer stuff not because I’m smarter but because I will sit there all day and night to figure something out while others will give up. I always thought that was my superpower in the job industry but now I’m not so sure if it will transfer to getting AI to do what I need done…
Same, I barely made it through Engineering school, but would stay up all night figuring out everything a computer could do (before the internet).
I did it because I enjoyed it, and still do. I just do it with LLMs now. There is more to figure out than ever before and things get created faster than I have time to understand them.
LLM should be enabling this, not making it more depressing.
But even tenacity is not enough. You also need an internal timer. "Wait a minute, this is taking too long, it shouldn't be this hard. Is my overall approach completely wrong?"
I'm not sure AIs have that. Humans do, or at least the good ones do. They don't quit on the problem, but they know when it's time to consider quitting on the approach.
> AIs have endless grit (or at least as endless as your budget).
That is the only thing he doesn't address: the money it costs to run the AI. If you let the agents loose, they easily burn north of 100M tokens per hour. Now at $25/1M tokens that gets quickly expensive. At some point, when we are all drug^W AI dependent, the VCs will start to cash in on their investments.
LLMs do not have grit or tenacity. Tenacity doesn't desribe a machine that doesn't need sleep or experience tiredness, or stress. Grit doesn't describe a chatbot that will tirelessly spew out answers and code because it has no stake or interest in the result, never perceives that it doesn't know something, and never reflects on its shortcomings.
> What happens to the "10X engineer" - the ratio of productivity between the mean and the max engineer? It's quite possible that this grows a lot.
I was thinking about this the other day as relates to the DevOps movement.
The DevOps movement started as a way to accelerate and improve the results of dev<->ops team dynamics. By changing practices and methods, you get acceleration and improvement. That creates "high-performing teams", which is the team form of a 10x engineer. Whether or not you believe in '10x engineers', a high-performing team is real. You really can make your team deploy faster, with fewer bugs. You have to change how you all work to accomplish it, though.
To get good at using AI for coding, you have to do the same thing: continuous improvement, changing workflows, different designs, development of trust through automation and validation. Just like DevOps, this requires learning brand new concepts, and changing how a whole team works. This didn't get adopted widely with DevOps because nobody wanted to learn new things or change how they work. So it's possible people won't adapt to the "better" way of using AI for coding, even if it would produce a 10x result.
If we want this new way of working to stick, it's going to require education, and a change of engineering culture.
This is an interesting thing that I'm contemplating. I also do believe that (perhaps with very few exceptions) there are no "10x engineers" by themselves, but engineers that thrive 10x more in a context or another (like, I'm sure Jeff Dean is an absolutely awesome engineer - but if you took him out of Google and plugged him into IBM - would he have had the same impact?)
With that in mind - I think one very unexplored area is "how to make the mixed AI-human teams successful". Like, I'm fairly convinced AI changes things, but to get to the industrialization of our craft (which is what management seems to want - and, TBH, something that makes sense from an economic pov), I feel that some big changes need to happen, and nobody is talking about that too much. What are the changes that need to happen? How do we change things, if we are to attempt such industrialization?
I'm pretty happy with Copilot in VS Code. Type what change I want Claude to make in the Copilot panel, and then use the VS Code in context diffs to accept or reject the proposed changes. While being able to make other small changes on my own.
So I think this tracks with Karpathy's defense of IDEs still being necessary ?
Has anyone found it practical to forgo IDEs almost entirely?
I've found copilot chat is able to do everything I need. I tried the Claude plugin for vscode and it was a noticeably worse experience for me.
Mind you copilot has only supported agent mode relatively recently.
I really like the way copilot does changes in such a way you can accept or reject and even revert to point in time in the chat history without using git. Something about this just fits right with how my brain works.
Using Claude plugin just felt like I had one hand tied behind my back.
I find Claude Code in VS Code is sometimes horribly inefficient. I tell it to replace some print-statements with proper logging in the one file I have open and it first starts burning tokens to understand the codebase for the 13th time today, despite not needing to and having it laid out in the CLAUDE.md already.
I have been assigning issues to copilot in Github. It will then create a pull request and work on and report back on the issue in the PR. I will pull the code and make small changes locally using VSCode when needed.
But what I like about this setup is that I have almost all the context I need to review the work in a single PR. And I can go back and revisit the PR if I ever run into issues down the line. Plus you can run sessions in parallel if needed, although I don't do that too much.
Are you letting it run your tests and run little snippets of code to try them out (like "python -c 'import module; print(module.something())'") or are you just using it to propose diffs for you to accept or reject?
This stuff gets a whole lot more interesting when you let it start making changes and testing them by itself.
> Tenacity. It's so interesting to watch an agent relentlessly work at something. They never get tired, they never get demoralized, they just keep going and trying things where a person would have given up long ago to fight another day.
This is true to an extent for sure and they will go much longer than most engineers without getting "tired", but I've def seen both sonnet and opus give up multiple times. They've updated code to skip tests they couldn't get to pass, given up on bugs they couldn't track down, etc. I literally had it ask "could we work on something else and come back to this"
The glorified autocomplete. Why would the LLM "work on something else then get back on this", is it's subconscious going to solve the problem during that time?
But because people say it, it says it too. Making sense is optional.
Ive found that clearing the context and getting back to it later actually DOES work. When you restart, your personal context is cleared and you might be better at describing the problem you are solving in a more informationally dense way.
LLM coding splits up engineers based on those who primarily like building and those who primarily like code reviews and quality assessment. I definitely don’t love the latter (especially when reviewing decisions not made by a human with whom I can build long-term personal rapport).
After certain experience threshold of making things from scratch, “coding” (never particularly liked that term) has always been 99% building, or architecture, and I struggle to see how often a well-architected solution today, with modern high-level abstractions, requires so much code that you’d save significant time and effort by not having to just type, possibly with basic deterministic autocomplete, exactly what you mean (especially considering you would have to also spend time and effort reviewing whatever was typed for you if you used a non-deterministic autocomplete).
"those who primarily like code reviews and quality assessment" -- I don't love those. In fact I find it tedious and love it when I can work on my own without them.
Except after 25 years of working I know how imperative they are, how easily a project can disintegrate into confused silos, and am frustrated as heck with these tools being pushed without attention to this problem.
See, I don't take it that extreme: LLMs make fantastic, never-before seen quality autocompletes. I hacked together a Neovim plugin that prompts an LLM to "finish this function" on command, and it's a big time save for the menial plumbing type operations. Think things like "this api I use expects JSON that encodes some subset of SQL, I want all the dogs with Ls in their name that were born on a Tuesday". Given an example of such API (or if the documentation ended up in its training), LLMs will consistently one-shot stuff like that.
Asking it to do entire projects? Dumb. You end up with spaghetti, unless you hand-hold it to a point that you might as well be using my autocomplete method.
Readit News
Like there have been multiple times now where I wanted the code to look a certain way, but it kept pulling back to the way it wanted to do things. Like if I had stated certain design goals recently it would adhere to them, but after a few iterations it would forget again and go back to its original approach, or mix the two, or whatever. Eventually it was easier just to quit fighting it and let it do things the way it wanted.
What I've seen is that after the initial dopamine rush of being able to do things that would have taken much longer manually, a few iterations of this kind of interaction has slowly led to a disillusionment of the whole project, as AI keeps pushing it in a direction I didn't want.
I think this is especially true if you're trying to experiment with new approaches to things. LLMs are, by definition, biased by what was in their training data. You can shock them out of it momentarily, whish is awesome for a few rounds, but over time the gravitational pull of what's already in their latent space becomes inescapable. (I picture it as working like a giant Sierpinski triangle).
I want to say the end result is very akin to doom scrolling. Doom tabbing? It's like, yeah I could be more creative with just a tad more effort, but the AI is already running and the bar to seeing what the AI will do next is so low, so....
This would be fine if not for one thing: the meta-skill of learning to use the LLM depreciates too. Today's LLM is gonna go away someday, the way you have to use it will change. You will be on a forever treadmill, always learning the vagaries of using the new shiny model (and paying for the privilege!)
I'm not going to make myself dependent, let myself atrophy, run on a treadmill forever, for something I happen to rent and can't keep. If I wanted a cheap high that I didn't mind being dependent on, there's more fun ones out there.
You're lucky to afford the luxury not to atrophy.
It's been almost 4 years since my last software job interview and I know the drills about preparing for one.
Long before LLMs my skills naturally atrophy in my day job.
I remember the good old days of J2ME of writing everything from scratch. Or writing some graph editor for universiry, or some speculative, huffman coding algorithm.
That kept me sharp.
But today I feel like I'm living in that netflix series about people being in Hell and the Devil tricking them they're in Heaven and tormenting them: how on planet Earth do I keep sharp with java, streams, virtual threads, rxjava, tuning the jvm, react, kafka, kafka streams, aws, k8s, helm, jenkins pipelines, CI-CD, ECR, istio issues, in-house service discovery, hierarchical multi-regions, metrics and monitoring, autoscaling, spot instances and multi-arch images, multi-az, reliable and scalable yet as cheap as possible, yet as cloud native as possible, hazelcast and distributed systems, low level postgresql performance tuning, apache iceberg, trino, various in-house frameworks and idioms over all of this? Oh, and let's not forget the business domain, coding standards, code reviews, mentorships and organazing technical events. Also, it's 2026 so nobody hires QA or scrum masters anymore so take on those hats as well.
So LLMs it is, the new reality.
I agree with the sentiment but I would have framed it differently. The LLM is a tool, just like code completion or a code generator. Right now we focus mainly on how to use a tool, the coding agent, to achieve a goal. This takes place at a strategic level. Prior to the inception of LLMs, we focused mainly on how to write code to achieve a goal. This took place at a tactical level, and required making decisions and paying attention to a multitude of details. With LLMs our focus shifts to a higher-level abstraction. Also, operational concerns change. When writing and maintaining code yourself, you focus on architectures that help you simplify some classes of changes. When using LLMs, your focus shifts to building context and aiding the model effectively implement their changes. The two goals seem related, but are radically different.
I think a fairer description is that with LLMs we stop exercising some skills that are only required or relevant if you are writing your code yourself. It's like driving with an automatic transmission vs manual transmission.
This is my fear - what happens if the AI companies can't find a path to profitability and shut down?
I haven’t found this to be true at all, at least so far.
As models improve I find that I can start dropping old tricks and techniques that were necessary to keep old models in line. Prompts get shorter with each new model improvement.
It’s not really a cycle where you’re re-learning all the time or the information becomes outdated. The same prompt structure techniques are usually portable across LLMs.
This isn't to say LLMs won't change software development forever, I think they will. But I doubt anyone has any idea what kind of tools and approaches everyone will be using 5 or 10 years from now, except that I really doubt it will be whatever is being hyped up at this exact moment.
My gripe with AI tools in the past is that the kind of work I do is large and complex and with previous models it just wasn't efficient to either provide enough context or deal with context rot when working on a large application - especially when that application doesn't have a million examples online.
I've been trying to implement a multiplayer game with server authoritative networking in Rust with Bevy. I specifically chose Bevy as the latest version was after Claude's cut off, it had a number of breaking changes, and there aren't a lot of deep examples online.
Overall it's going well, but one downside is that I don't really understand the code "in my bones". If you told me tomorrow that I had optimize latency or if there was a 1 in 100 edge case, not only would I not know where to look, I don't think I could tell you how the game engine works.
In the past, I could not have ever gotten this far without really understanding my tools. Today, I have a semi functional game and, truth be told, I don't even know what an ECS is and what advantages it provides. I really consider this a huge problem: if I had to maintain this in production, if there was a SEV0 bug, am I confident enough I could fix it? Or am I confident the model could figure it out? Or is the model good enough that it could scan the entire code base and intuit a solution? One of these three questions have to be answered or else brain atrophy is a real risk.
I am interested in doing something similar (Bevy. not multiplayer).
I had the thought that you ought be able to provide a cargo doc or rust-analyzer equivalent over MCP? This... must exist?
I'm also curious how you test if the game is, um... fun? Maybe it doesn't apply so much for a multiplayer game, I'm thinking of stuff like the enemy patterns and timings in a soulslike, Zelda, etc.
I did use ChatGPT to get some rendering code for a retro RCT/SimCity-style terrain mesh in Bevy and it basically worked, though several times I had to tell it "yeah uh nothing shows up", at which point is said "of course! the problem is..." and then I learned about mesh winding, fine, okay... felt like I was in over my head and decided to go to a 2D game instead so didn't pursue that further.
It cuts down the input tokens significantly which is nice for the monthly bill, but I found the main benefit is that it actually stops the model from getting distracted by existing implementation details. It feels a bit like overengineering but it makes reasoning about the system architecture much more reliable when you don't have to dump the whole codebase into the context window.
Deleted Comment
Man, I absolutely hate this feeling.
Its what is known as one of the Socratic "myths," and really just contributes to a web of concepts that leads the dialogue to its ultimate terminus of aporia (being a relatively early Plato dialogue). Socrates, characteristically, doesn't really give his take on writing. In the text, he is just trying to help his friend write a horny love letter/speech!
I can't bring it up right now, but the end of the dialogue has a rather beautiful characterization of writing in the positive, saying that perhaps logos can grow out of writing, like a garden.
I think if pressed Socrates/Plato would say that LLM's are merely doxa machines, incapable of logos. But I am just spitballing.
Thankfully more and more people are seriously considering the effects of technology on true wisdom and getting of the "all technological progress clearly is great, look at all these silly unenlightened naysayers from the past" train.
To me, this feels like support. I was never an adult who could not read or write, so I can't check my experience against Socrates' specific concern. But speaking to the idea of memory, I now "outsource" a lot of my memory to my smartphone.
In the past, I would just remember my shopping list, and go to the grocery store and get what I needed. Sure, sometimes I'd forget a thing or two, but it was almost always something unimportant, and rarely was a problem. Now I have my list on my phone, but on many occasions where I don't make a shopping list on my phone, when I get to the grocery store I have a lot of trouble remembering what to get, and sometimes finish shopping, check out, and leave the store, only to suddenly remember something important, and have to go back in.
I don't remember phone numbers anymore. In college (~2000) I had the campus numbers (we didn't have cell phones yet) of at least two dozen friends memorized. Today I know my phone number, my wife's, and my sister's, and that's it. (But I still remember the phone number for the first house I lived in, and we moved out of that house when I was five years old. Interestingly, I don't remember the area code, but I suppose that makes sense, as area codes weren't required for local dialing in the US back in the 80s.)
Now, some of this I will probably ascribe to age: I expect our memory gets more fallible as we get older (I'm in my mid 40s). I used to have all my credit/debit card numbers, and their expiration dates and security codes, memorized (five or six of them), but nowadays I can only manage to remember two of them. (And I usually forget or mix up the expiration dates; fortunately many payment forms don't seem to check, or are lax about it.) But maybe that is due to new technology to some extent: most/all sites where I spend money frequently remember my card for me (and at most only require me to enter the security code). And many also take Paypal or Google Pay, which saves me from having to recall the numbers.
So I think new technology making us "dumber" is a very real thing. I'm not sure if it's a good thing or a bad thing. You could say that, in all of my examples, technology serving the place of memory has freed up mental cycles to remember more important things, so it's a net positive. But I'm not so sure.
My personal counterpoint is Norman's thesis in Things That Make Us Smart.
I've long tried, and mostly failed, to consider the tradeoffs, to be ever mindful that technologies are never neutral (winners & losers), per Postman's Technopoly.
In all seriousness though, it's just crazy that anybody is thinking these things at the dawn of civilization.
And "other people in the past predicted doom about something like this and it didn't happen" is a fallacious non-argument even when the things are comparable.
Using an LLM is almost exactly the same. You get the occasional, "wow! I've never seen it do that before!" moments (whether that thing it just did was even useful or not), get a short hit of feel goods, and then we keep using it trying to get another hit. It keeps providing them at just the right intervals for people to keep them going just like they do with tick tock
https://www.fast.ai/posts/2026-01-28-dark-flow/
As in if the LLM doesn't know about it, some devs are basically giving up and not even going to RTFM. I literally had to explain to someone today how something works by...reading through the docs and linking them the docs with screenshots and highlighted paragraphs of text.
Still got push back along the lines of "not sure if this will work". It's. Literally. In. The. Docs.
15 years ago I was working in an environment where they had lots of Indians as cheap labour - and the same thing will show up in any environment where you go for hiring a mass of cheap people while looking more at the cost than at qualifications: You pretty much need to trick them into reading stuff that are relevant.
I remember one case where one had a problem they couldn't solve, and couldn't give me enough info to help remotely. In the end I was sitting next to them, and made them read anything showing up on the screen out loud. Took a few tries where they were just closing dialog boxes without reading it, but eventually we had that under control enough that they were able to read the error messages to me, and then went "Oh, so _that's_ the problem?!"
Overall interacting with a LLM feels a lot like interacting with one of them back then, even down to the same excuses ("I didn't break anything in that commit, that test case was never passing") - and my expectation for what I can get out of it is pretty much the same as back then, and approach to interacting with it is pretty similar. It's pretty much an even cheaper unskilled developer, you just need to treat it as such. And you don't pair it up with other unskilled developers.
You can have as many extremely detailed and easy to parse gudies, references, etc. there will always be a portion of customers who refuse to read them.
Never could figure out why because they aren't stupid or anything.
I wouldn't have believed it a few tears ago if you told me the industry would one day, in lockstep, decide that shipping more tech-debt is awesome. If the unstated bet doesn't pay off, that is, AI development will outpace the rate it generates cruft, then there will be hell to pay.
Once we realize the kind of mess _those_ models created, well, we'll need even more capable models.
It's a variation on the theme of Kernighan insight about the more "clever" you are while coding the harder it will be to debug.
EDIT: Simplicity is a way out but it's hard under normal circumstances, now with this kind of pressure to ship fast because the colleague with the AI chimp can outperform you, aiming at simplicity will require some widespread understanding
This isn't anything new of course. Previously it was with projects built by looking for the cheapest bidder and letting them loose on an ill-defined problem. And you can just imagine what kind of code that produced. Except the scale is much larger.
My favorite example of this was a project that simply stopped working due to the amount of bugs generated from layers upon layers of bad code that was never addressed. That took around 2 years of work to undo. Roughly 6 months to un-break all the functionality and 6 more months to clean up the core and then start building on top.
Brilliant. Even if it was a typo.
And guess what, I'm finally convinced they're right.
Consider: it's been that way for decades. We may tell ourselves good developers write quality code given the chance, but the truth is, the median programmer is a junior with <5 years of experience, and they cannot write quality code to save their life. That's purely the consequence of rapid growth of software industry itself. ~all production code in the past few decades was written by juniors, it continues to be so today; those who advance to senior level end up mostly tutoring new juniors instead of coding.
Or, all that put another way: tech debt is not wrong. It's a tool, a trade-off. It's perfectly fine to be loaded with it, if taking it lets you move forward and earn enough to afford paying installments when they're due. Like with housing: you're better off buying it with lump payment, or off savings in treasury bonds, but few have that money on hand and life is finite, so people just get a mortgage and move on.
--
Edited to add: There's a silver lining, though. LLMs make tech debt legible and quantifiable.
LLMs are affected by tech debt even more than human devs are, because (currently) they're dumber, they have less cognitive capability around abstractions and generalizations[0]. They make up for it by working much faster - which is a curse in terms of amplifying tech debt, but also a blessing, because you can literally see them slowing down.
Developer productivity is hard to measure in large part because the process is invisible (happens in people's heads and notes), and cause-and-effect chains play out over weeks or months. LLM agents compress that to hours to days, and the process itself is laid bare in the chat transcript, easy to inspect and analyze.
The way I see it, LLMs will finally allow us to turn software development at tactical level from art into an engineering process. Though it might be too late for it to be of any use to human devs.
--
[0] - At least the out-of-distribution ones - quirks unique to particular codebase and people behind it.
It's not debt if you never have to pay it back. If a model can regenerate a whole relibale codebase in minutes from a spec, then your assessment of "tech debt" in that output becomes meaningless.
(except where it's been stated, championed, enforced, and ultimated in no unequivocal terms by every executive in the tech industry)
I think the way you’re using these tools that makes you feel this way is a choice. You’re choosing to not be in control and do as little as possible.
They are amazing for side projects but not for serious code with real world impact where most of the context is in multiple people's head.
At some point, I find myself needing to disconnect out of overwhelm and frustration. Faster responses isn't necessarily better. I want more observability in the development process so that I can be a party to it. I really have felt that I need to orchestrate multiple agents working in tandem, playing sort of a bad-cop, good-cop and a maybe a third trying to moderate that discussion and get a fourth to effectively incorporate a human in the mix. But that's too much to integrate in my day job.
I found the setting and turned it off for real. Good riddance. I’ll use the hotkey on occasion.
I use claude daily, no problems with it. But vscode + copilot suggestions was garbage!
Now back to IC with 25+ years of experience + LLM = god mode, and its fun again.
Absolutely. At a certain level of usage, you just have to let it do it's thing.
People are going to take issue with that. You absolutely don't have to let it do its thing. In that case you have to be way more in the loop. Which isn't necessarily a bad thing.
But assuming you want it to basically do everything while you direct it, it becomes pointless to manage certain details. One thing in my experience is that Claude always wants to use ReactRouter. My personal preference is TanStack router, so I asked it to use it initially. That never really created any problems but after like the 3rd time of realizing I forget to specify it, I also realized that it's totally pointless. ReactRouter works fine and Claude uses it fine - its pointless to specify otherwise.
Yea exactly, Like we are just waiting so that it gets completed and after it gets completed then what? We ask it to do new things again.
Just as how if we are doom scrolling, we watch something for a minute then scroll down and watch something new again.
The whole notion of progress feels completely fake with this. Somehow I guess I was in a bubble of time where I had always end up using AI in web browsers (just as when chatgpt 3 came) and my workflow didn't change because it was free but recently changed it when some new free services dropped.
"Doom-tabbing" or complete out of the loop AI agentic programming just feels really weird to me sucking the joy & I wouldn't even consider myself a guy particular interested in writing code as I had been using AI to write code for a long time.
I think the problem for me was that I always considered myself a computer tinker before coder. So when AI came for coding, my tinkering skills were given a boost (I could make projects of curiosity I couldn't earlier) but now with AI agents in this autonomous esque way, it has come for my tinkering & I do feel replaced or just feel like my ability of tinkering and my interests and my knowledge and my experience is just not taken up into account if AI agent will write the whole code in multi file structure, run commands and then deploy it straight to a website.
I mean my point is tinkering was an active hobby, now its becoming a passive hobby, doom-tinkering? I feel like I have caught up on the feeling a bit earlier with just vibe from my heart but is it just me who feels this or?
What could be a name for what I feel?
A quick example is trying to build a simple expenses app with it. I just want to store a list of transactions with it. I’ve already written the types and data model and just need the AI to give me the plumping. And it will always end up inserting recommendations about double entry bookkeeping.
Not trusting the ML's output is step one here, that keeps you intellectually involved - but it's still a far cry from solving the majority of problems yourself (instead you only solve problems ML did a poor job at).
Step two: I delineate interesting and uninteresting work, and Claude becomes a pair programmer without keyboard access for the latter - I bounce ideas off of it etc. making it an intelligent rubber duck. [Edit to clarify, a caveat is that] I do not bore myself with trivialities such as retrieving a customer from the DB in a REST call (but again, I do verify the output).
Genuine question, why isn't your ORM doing that? I see a lot of use cases for LLMs that seem to be more expensive ways to do snippets and frameworks...
I used to know a person like that - high in the company structure who would claim he was a great engineer, but all the actual engineers would make jokes about him and his ancient skills during private conversations.
The time it happened for me was rather abrupt, with no training in between, and the feeling was eerily similar.
You know _exactly_ why the best solution is, you talk to your reports, but they have minds of their own, as well as egos, and they do things … their own way.
At some point I stopped obsessing with details and was just giving guidance and direction only in the cases where it really mattered, or when asked, but let people make their own mistakes.
Now LLMs don’t really learn on their own or anything, but the feeling of “letting go of small trivial things” is sorta similar. You concentrate on the bigger picture, and if it chose to do an iterative for loop instead of using a functional approach the way you like it … well the tests still pass, don’t they.
You then tell your agent to always run that skill prior to moving on. If the examples are pattern matchable you can even have the agent write custom lints if your linter supports extension or even write a poor man’s linter using ast-grep.
I usually have a second session running that is mainly there to audit the code and help me add and adjust skills while I keep the main session on the task of working on the feature. I've found this far easier to stay engaged than context switching between unrelated tasks.
Context management, proper prompting and clear instructions, proper documentation are still relevant.
However, for hobby projects where I purposely use tech I’m not very familiar with, I force myself not to use LLMs at all - even as a chat. Thus, operating The old way - writing code manually, reading documentation, etc brings me a joy of learning back and, hopefully, establishes new neurone connections.
I would argue this is ok for front-end. For back-end? very, very bad- if you can't get a usable output do it by hand.
[1] - https://openspec.dev/
The AI definitely has preferences and attention issues, but there are ways to overcome them.
Defining code styles in a design doc, and setting up initial examples in key files goes a long way. Claude seems pretty happy to follow existing patterns under these conditions unless context is strained.
I have pretty good results using a structured workflow that runs a core loop of steps on each change, with a hook that injects instructions to keep attention focused.
In the happy case where I have a good idea of the changes necessary, I will ask it to do small things, step by step, and examine what it does and commit.
In the unhappy case where one is faced with a massive codebase and no idea where to start, I find asking it to just “do the thing” generates slop, but enough for me to use as inspiration for the above.
Dead Comment
Dead Comment
Have to really look out for the crap.
I’ve always said I’m a builder even though I’ve also enjoyed programming (but for an outcome, never for the sake of the code)
This perfectly sums up what I’ve been observing between people like me (builders) who are ecstatic about this new world and programmers who talk about the craft of programming, sometimes butting heads.
One viewpoint isn’t necessarily more valid, just a difference of wiring.
"I got into programming because I like programming, not whatever this is..."
Yes, I'm building stupid things faster, but I didn't get into programming because I wanted to build tons of things. I got into it for the thrill of defining a problem in terms of data structures and instructions a computer could understand, entering those instructions into the computer, and then watching victoriously while those instructions were executed.
If I was intellectually excited about telling something to do this for me, I'd have gotten into management.
I used Claude Code to implement a OpenAI 4o-vision powered receipt scanning feature in an expense tracking tool I wrote by hand four years ago. It did it in two or three shots while taking my codebase into account.
It was very neat, and it works great [^0], but I can't latch onto the idea of writing code this way. Powering through bugs while implementing a new library or learning how to optimize my test suite in a new language is thrilling.
Unfortunately (for me), it's not hard at all to see how the "builders" that see code as a means to an end would LOVE this, and businesses want builders, not crafters.
In effect, knowing the fundamentals is getting devalued at a rate I've never seen before.
[^0] Before I used Claude to implement this feature, my workflow for processing receipts looked like this: Tap iOS Shortcut, enter the amount, snap a pic of the receipt, type up the merchant, amount and description for the expense, then have the shortcut POST that to my expenses tracking toolkit which, then, POSTs that into a Google Sheet. This feature amounted the need for me to enter the merchant and amount. Unfortunately, it often took more time to confirm that the merchant, amount and date details OpenAI provided were correct (and correct it when details were wrong, which was most of the the time) than it did to type out those details manually, so I just went back to my manual workflow. However, the temptation to just glance at the details and tap "This looks correct" was extremely high, even if the info it generated was completely wrong! It's the perfect analogue to what I've been witnessing throughout the rise of the LLMs.
>If I was intellectually excited about telling something to do this for me, I'd have gotten into management.
Exactly this. This is the simplest and tersest way of explaining it yet.
There is a strange insistence on not helping the LLM arrive at the best outcome in the subtext to this question a lot of times. I feel like we are living through the John Henry legend in real time
You can still do that with Claude Code. In fact, Claude Code works best the more granular your instructions get.
So maybe our common ground is that we are direct problem solvers. :-)
What I mean by that: you had compiled vs interpreted languages, you had types vs untyped, testing strategies, all that, at least in some part, was a conversation about the tradeoffs between moving fast/shipping and maintainability.
But it isn't just tech, it is also in methodologies and the words use, from "build fast and break things" and "yagni" to "design patterns" and "abstractions"
As you say, it is a different viewpoint... but my biggest concern with where are as industry is that these are not just "equally valid" viewpoints of how to build software... it is quite literally different stages of software, that, AFAICT, pretty much all successful software has to go through.
Much of my career has been spent in teams at companies with products that are undergoing the transition from "hip app built by scrappy team" to "profitable, reliable software" and it is painful. Going from something where you have 5 people who know all the ins and outs and can fix serious bugs or ship features in a few days to something that has easy clean boundaries to scale to 100 engineers of a wide range of familiarities with the tech, the problem domain, skill levels, and opinions is just really hard. I am not convinced yet that AI will solve the problem, and I am also unsure it doesn't risk making it worse (at least in the short term)
Much of my career has been spent in teams at companies with products that are undergoing the transition from "hip app built by scrappy team" to "profitable, reliable software" and it is painful. Going from something where you have 5 people who know all the ins and outs and can fix serious bugs or ship features in a few days to something that has easy clean boundaries to scale to 100 engineers of a wide range of familiarities with the tech, the problem domain, skill levels, and opinions is just really hard. I am not convinced yet that AI will solve the problem, and I am also unsure it doesn't risk making it worse (at least in the short term)
“””
This perspective is crucial. Scale is the great equalizer / demoralizer, scale of the org and scale of the systems. Systems become complex quickly, and verifiability of correctness and function becomes harder. Companies that built from day with AI and have AI influencing them as they scale, where does complexity begin to run up against the limitations of AI and cause regression? Or if all goes well, amplification?
I test all of the code I produce via LLMs, usually doing fairly tight cycles. I also review the unit test coverage manually, so that I have a decent sense that it really is testing things - the goal is less perfect unit tests and more just quickly catching regressions. If I have a lot of complex workflows that need testing, I'll have it write unit tests and spell out the specific edge cases I'm worried about, or setup cheat codes I can invoke to test those workflows out in the UI/CLI.
Trust comes from using them often - you get a feeling for what a model is good and bad at, and what LLMs in general are good and bad at. Most of them are a bit of a mess when it comes to UI design, for instance, but they can throw together a perfectly serviceable "About This" HTML page. Any long-form text they write (such as that About page) is probably trash, but that's super-easy to edit manually. You can often just edit down what they write: they're actually decent writers, just very verbose and unfocused.
I find it similar to management: you have to learn how each employee works. Unless you're in the Top 1%, you can't rely on every employee giving 110% and always producing perfect PRs. Bugs happen, and even NASA-strictness doesn't bring that down to zero.
And just like management, some models are going to be the wrong employee for you because they think your style guide is stupid and keep writing code how they think it should be written.
And accountability can still exist? Is the engineer that created or reviewed a Pull Request using Claude Code less accountable then one that used PICO?
[1] https://www.klio.org/theory-of-llm-dev-skepticism/
Took me a few years to realize that this wasn't a universal feeling, and that many others found the programming tasks more fulfilling than any challenging engineering. I suppose this is merely another manifestation of the same phenomena.
The fact of the matter is LLMs produce lower quality at higher volumes in more time than it would take to write it myself, and I’m a very mediocre engineer.
I find this seperation of “coding” vs “building” so offensive. It’s basically just saying some people are only concerned with “inputs”, while others with “outputs”. This kind of rhetoric is so toxic.
It’s like saying LLM art is separating people into people who like to scribble, and people who like to make art.
This distinction to me separates the two primary camps
> I’ve always said I’m a builder even though I’ve also enjoyed programming (but for an outcome, never for the sake of the code)
> This perfectly sums up what I’ve been observing between people like me (builders) who are ecstatic about this new world and programmers who talk about the craft of programming, sometimes butting heads.
That's one take, sure, but it's a specially crafted one to make you feel good about your position in this argument.
The counter-argument is that LLM coding splits up engineers based on those who primarily like engineering and those who like managing.
You're obviously one of the latter. I, OTOH, prefer engineering.
It's just the level of engineering we're split on. I like the type of engineering where I figure out the flow of data, maybe the data structures and how they move through the system.
Writing the code to do that is the most boring part of my job. The LLM does it now. I _know_ how to do it, I just don't want to.
It all boils down to communication in a way. Can you communicate what you want in a way others (in this case a language model) understands? And the parts you can't communicate in a human language, can you use tools to define those (linters, formatters, editorconfig)?
I've done all that with actual humans for ... a decade? So applying the exact same thing to a machine is weirdly more efficient, it doesn't complain about the way I like to have my curly braces - it just copies the defined style. With humans I've found out that using impersonal tooling to inspect code style and flaws has a lot less friction than complaining about it in PR reviews. If the CI computer says no, people don't complain, they fix it.
This is much less significant than the fact LLMs split engineers on those who primarily like quality v. those who primarily like speed.
'Coders' make 'builders' keep the source code good enough so that 'builders' can continue building without breaking what they built.
If 'builders' become x10 productive and 'coders' become unable to keep up with unsurmountable pile of unmaintainable mess that 'builders' proudly churn out, 'bullders' will start to run into impossibility to build further without starting over and over again hoping that agents will be able to get it right this time.
Then force the builders to use those tools to constrain their output.
I think both approaches are okay, the biggest thing for me is the former needs to test way more, and review the code more, as developers we don't read code enough, with the "prompt and forget" approach we have a lot of free time we could spend reading the code, asking the model to refactor and refine the code. I am shocked when I hear about hundreds of thousands of lines in some projects. I've rebuilt Beads from the ground up and I'm under 10 lines of code.
So we're going to have various level of AI Code Builders if you will: Junior, Mid, Senior, Architect. I don't know if models will ever pick up the slack for Juniors any time soon. We would need massive context windows for models, and who will pay for that? We need a major AI breakthrough to where the cost goes down drastically before that becomes profitable.
We have services deployed globally serving millions of customers where rigor is really important.
And we have internal users who're building browser extensions with AI that provide valuable information about the interface they're looking at including links to the internal record management, and key metadata that's affecting content placement.
These tools could be handed out on Zip drives in the street and it would just show our users some of the metadata already being served up to them, but it's amazing to strip out 75% of the process of certain things and just have our user (in this case though, it's one user who is driving all of this, so it does take some technical inclination) build out these tools that save our editors so much time when doing this before would have been months and months and months of discovery and coordination and designs that probably wouldn't actually be as useful in the end after the wants of the user are diluted through 18 layers of process.
Managers and project managers are valuable roles and have important skill sets. But there's really very little connection with the role of software development that used to exist.
It's a bit odd to me to include both of these roles under a single label of "builders", as they have so little in common.
EDIT: this goes into more detail about how coding (and soon other kinds of knowledge work) is just a management task now: https://www.oneusefulthing.org/p/management-as-ai-superpower...
I deliberately avoid full vibe coding since I think doing so will rust my skills as a programmer. It also really doesn’t save much time in my experience. Once I have a design in mind, implementation is not the hard part.
For a long time in my career now I've been in a situation where I'd be able to build more if I was willing to abstract myself and become a slide-merchant/coalition-builder. I don't want to do this though.
Yet, I'm still quite an enthusiastic vibe-coder.
I think it's less about coding Vs building and more about tolerance for abstraction and politics. And I don't think there are that many people who are so intolerant of abstraction that they won't let agents write a bunch of code for them.
I’ve always considered myself a “process” person, I would even get hung-up on certain projects because I enjoyed crafting them so much.
LLM’s have taken a bit of that “process” enjoyment from me, but I think have also forced some more “outcome” thinking into my head, which I’m taking as a positive.
I had felt like this and still do but man, at some point, I feel like the management churn feels real & I just feel suffering from a new problem.
Suppose, I actually end up having services literally deployed from a single prompt nothing else. Earlier I used to have AI write code but I was interested in the deployment and everything around it, now there are services which do that really neatly for you (I also really didn't give into the agent hype and mostly used browsers LLM)
Like on one hand you feel more free to build projects but the whole joy of project completely got reduced.
I mean, I guess I am one of the junior dev's so to me AI writing code on topics I didn't know/prototyping felt awesome.
I mean I was still involved in say copy pasting or looking at the code it generates. Seeing the errors and sometimes trying things out myself. If AI is doing all that too, idk
For some reason, recently I have been disinterested in AI. I have used it quite a lot for prototyping but I feel like this complete out of the loop programming just very off to me with recent services.
I also feel like there is this sense of if I buy for some AI thing, to maximally extract "value" out of it.
I guess the issue could be that I can have vague terms or have a very small text file as input (like just do X alternative in Y lang) and I am now unable to understand the architectural decisions and the overwhelmed-ness out of it.
Probably gonna take either spec-driven development where I clearly define the architecture or development where I saw something primagen do recently which is that the AI will only manipulate code of that particular function, (I am imagining it for a file as well) and somehow I feel like its something that I could enjoy more because right now it feels like I don't know what I have built at times.
When I prototype with single file projects using say browser for funsies/any idea. I get some idea of what the code kind of uses with its dependencies and functions names from start/end even if I didn't look at the middle
A bit of ramble I guess but the thing which kind of is making me feel this is that I was talking to somebody and shwocasing them some service where AI + server is there and they asked for something in a prompt and I wrote it. Then I let it do its job but I was also thinking how I would architect it (it was some detect food and then find BMR, and I was thinking first to use any api but then I thought that meh it might be hard, why not use AI vision models, okay what's the best, gemini seems good/cheap)
and I went to the coding thing to see what it did and it actually went even beyond by using the free tier of gemini (which I guess didn't end up working could be some rate limit of my own key but honestly it would've been the thing I would've tried too)
So like, I used to pride myself on the architectural decisions I make even if AI could write code faster but now that is taken away as well.
I really don't want to read AI code so much so honestly at this point, I might as well write code myself and learn hands on but I have a problem with build fast in public like attitude that I have & just not finding it fun.
I feel like I should do a more active job in my projects & I am really just figuring out what's the perfect way to use AI in such contexts & when to use how much.
Thoughts?
This is a mix of the “in the future, everyone will have a 3D printer at home and just 3D print random parts they need” and “anyone can trivially build Dropbox with rsync themselves” arguments.
Tech savvy users who know how to use LLMs aren’t how vendors of small utilities stay in business.
They stay in business because they sell things to users who are truly clueless with tech (99% of the population, which can’t even figure out the settings app on their phone), and solid distribution/marketing is how you reach those users and can’t really be trivially hacked because everyone is trying to hack it.
Or they stay in business because they offer some sort of guarantee (whether legal, technical, or other) that the users don’t want to burden themselves with because they have other, more important stuff to worry about.
Definitely. Making small, single-purpose utilities with LLMs is almost as easy these days as googling for them on-line - much easier, in fact, if you account for time spent filtering out all the malware, adware, "to finish the process, register an account" and plain broken "tools" that dominate SERP.
Case in point, last time my wife needed to generate a few QR codes for some printouts for an NGO event, I just had LLM make one as a static, single-page client-side tool and hosted it myself -- because that was the fastest way to guarantee it's fast, reliable, free of surveillance economy bullshit, and doesn't employ URL shorteners (surprisingly common pattern that sometimes becomes a nasty problem down the line; see e.g. a high-profile case of some QR codes on food products leading to porn sites after shortlink got recycled).
There has been a lot of research that shows that grit is far more correlated to success than intelligence. This is an interesting way to show something similar.
AIs have endless grit (or at least as endless as your budget). They may outperform us simply because they don't ever get tired and give up.
Full quote for context:
Tenacity. It's so interesting to watch an agent relentlessly work at something. They never get tired, they never get demoralized, they just keep going and trying things where a person would have given up long ago to fight another day. It's a "feel the AGI" moment to watch it struggle with something for a long time just to come out victorious 30 minutes later. You realize that stamina is a core bottleneck to work and that with LLMs in hand it has been dramatically increased.
"Listen, and understand! That Terminator is out there! It can't be bargained with. It can't be reasoned with. It doesn't feel pity, or remorse, or fear. And it absolutely will not stop... ever, until you are dead!"
Sometimes it's a
or a"this feature would require extensive logic and changes to the existing codebase".
Sometimes they just declare their work done. Ignoring failing tests and builds.
You can nudge them to keep going but I often feel like, when they behave like this, they are at their limit of what they can achieve.
If you try to single shot something perhaps. But with multiple shots, or an agent swarm where one agent tells another to try again, it'll keep going until it has a working solution.
Context matters, for an LLM just like a person. When I wrote code I'd add TODOs because we cannot context switch to another problem we see every time.
But you can keep the agent fixated on the task AND have it create these TODOs, but ultimately it is your responsibility to find them and fix them (with another agent).
I did it because I enjoyed it, and still do. I just do it with LLMs now. There is more to figure out than ever before and things get created faster than I have time to understand them.
LLM should be enabling this, not making it more depressing.
I'm not sure AIs have that. Humans do, or at least the good ones do. They don't quit on the problem, but they know when it's time to consider quitting on the approach.
That is the only thing he doesn't address: the money it costs to run the AI. If you let the agents loose, they easily burn north of 100M tokens per hour. Now at $25/1M tokens that gets quickly expensive. At some point, when we are all drug^W AI dependent, the VCs will start to cash in on their investments.
Dead Comment
I was thinking about this the other day as relates to the DevOps movement.
The DevOps movement started as a way to accelerate and improve the results of dev<->ops team dynamics. By changing practices and methods, you get acceleration and improvement. That creates "high-performing teams", which is the team form of a 10x engineer. Whether or not you believe in '10x engineers', a high-performing team is real. You really can make your team deploy faster, with fewer bugs. You have to change how you all work to accomplish it, though.
To get good at using AI for coding, you have to do the same thing: continuous improvement, changing workflows, different designs, development of trust through automation and validation. Just like DevOps, this requires learning brand new concepts, and changing how a whole team works. This didn't get adopted widely with DevOps because nobody wanted to learn new things or change how they work. So it's possible people won't adapt to the "better" way of using AI for coding, even if it would produce a 10x result.
If we want this new way of working to stick, it's going to require education, and a change of engineering culture.
With that in mind - I think one very unexplored area is "how to make the mixed AI-human teams successful". Like, I'm fairly convinced AI changes things, but to get to the industrialization of our craft (which is what management seems to want - and, TBH, something that makes sense from an economic pov), I feel that some big changes need to happen, and nobody is talking about that too much. What are the changes that need to happen? How do we change things, if we are to attempt such industrialization?
So I think this tracks with Karpathy's defense of IDEs still being necessary ?
Has anyone found it practical to forgo IDEs almost entirely?
Mind you copilot has only supported agent mode relatively recently.
I really like the way copilot does changes in such a way you can accept or reject and even revert to point in time in the chat history without using git. Something about this just fits right with how my brain works. Using Claude plugin just felt like I had one hand tied behind my back.
But what I like about this setup is that I have almost all the context I need to review the work in a single PR. And I can go back and revisit the PR if I ever run into issues down the line. Plus you can run sessions in parallel if needed, although I don't do that too much.
This stuff gets a whole lot more interesting when you let it start making changes and testing them by itself.
This is true to an extent for sure and they will go much longer than most engineers without getting "tired", but I've def seen both sonnet and opus give up multiple times. They've updated code to skip tests they couldn't get to pass, given up on bugs they couldn't track down, etc. I literally had it ask "could we work on something else and come back to this"
But because people say it, it says it too. Making sense is optional.
And then someone comes and "improves" their agent with additional "do not repeat yourself" prompts scattered all over the place, to no avail.
"Asinine" describes my experience perfectly.
After certain experience threshold of making things from scratch, “coding” (never particularly liked that term) has always been 99% building, or architecture, and I struggle to see how often a well-architected solution today, with modern high-level abstractions, requires so much code that you’d save significant time and effort by not having to just type, possibly with basic deterministic autocomplete, exactly what you mean (especially considering you would have to also spend time and effort reviewing whatever was typed for you if you used a non-deterministic autocomplete).
Except after 25 years of working I know how imperative they are, how easily a project can disintegrate into confused silos, and am frustrated as heck with these tools being pushed without attention to this problem.
Asking it to do entire projects? Dumb. You end up with spaghetti, unless you hand-hold it to a point that you might as well be using my autocomplete method.