We mourn our craft

I started programming over 40 years ago because it felt like computers were magic. They feel more magic today than ever before. We're literally living in the 1980s fantasy where you could talk to your computer and it had a personality. I can't believe it's actually happening, and I've never had more fun computing.

I can't empathize with the complaint that we've "lost something" at all. We're on the precipice of something incredible. That's not to say there aren't downsides (WOPR almost killed everyone after all), but we're definitely in a golden age of computing.

LLMs are only a threat if you see your job as a code monkey. In that case you're likely already obsoleted by outsourced staff who can do your job much cheaper.

If you see your job as a "thinking about what code to write (or not)" monkey, then you're safe. I expect most seniors and above to be in this position, and LLMs are absolutely not replacing you here - they can augment you in certain situations.

The perks of a senior is also knowing when not to use an LLM and how they can fail; at this point I feel like I have a pretty good idea of what is safe to outsource to an LLM and what to keep for a human. Offloading the LLM-safe stuff frees up your time to focus on the LLM-unsafe stuff (or just chill and enjoy the free time).

Agree with the author. I like the process of writing code, typing method names and class definitions while at the same time thinking ahead about overall architecture, structure, how much time given function would run for, what kind of tests are necessary.

I find it unsettling how many people in the comments say that they don't like writing code. Feels aliens to me. We went into this field for seemingly very different reasons.

I do use LLMs and even these past two days I was doing vibe coding project which was noticeably faster to setup and get to its current state than if I wrote in myself. However I feel almost dirty by how little I understand the project. Sure, I know the overall structure, decisions and plan. But I didn't write any of it and I don't have deep understanding of the codebase which I usually have when working on codebase myself.

While I'm on the fence about LLMs there's something funny about seeing an industry of technologists tear their own hair out about how technology is destroying their jobs. We're the industry of "we'll automate your job away". Why are we so indignant when we do it to ourselves...

> If you would like to grieve, I invite you to grieve with me.

I’m not against AI in itself, but the current implementation (read market) can eat my shiny metal … I got kids to feed and bills to pay. Yes AI was all about post scarcity. Hiring in our industry is already feeling the brunt of AI, where is the UBI? Where is our soft landing? I’m too old to learn XYZ skill if I get laid off, it will be life changing, and for what? For some rich person to be richer while simultaneously destroying the climate and my energy bill?

I would push back, but I do not know how. Hope for the market to pop I guess.

These comments are comical. How hard is it to understand that human beings are experiential creatures. Our experiences matter, to survival, to culture, and identity.

I mourn the horse masters and stable boys of a century past because of their craft. Years of intuition and experience.

Why do you watch a chess master play, or a live concert, or any form of human creation?

Should we automate parts of our profession? Yes.

Should he mourn the loss of our craft. Also yes.

The craft is still there just like painting is still an alternative to a photograph. It's just not going to be valued by society anymore, and far fewer will learn how to do it. Natural language is the new programming language. For now, understanding the craft is still an edge in making better prompts, but already I can see that telling Antigravity "now look for ways to make this more efficient" works almost as well as guiding it specifically on how it had duplicated some code flows.

I do feel a kind of personal loss in the sense that society is in the process of stopping to value or admire the design and coding skill I've cultivated since I was 6yo. At the same time, I'm kind of thrilled that I can write a detailed readme.md and tell an agent to "make it so" and I can iterate to a utility program in 20 minutes instead of an hour. When I feel a pit in my stomach is when that utility program uses some framework that I haven't learned, and don't need to because their code worked perfectly the first time. Surely that means I'm going to basically stop learning the details, as the details I've accumulated over my life quickly begin to not matter anymore.

Honestly I'm planning to use AI to make a kick-ass retro development environment a la "Sending Modern Languages Back to 1980s Game Programmers" (https://prog21.dadgum.com/6.html) and spend my retirement having fun in it.

I do not mourn.

For my whole life I’ve been trying to make things—beautiful elegant things.

When I was a child, I found a cracked version of Photoshop and made images which seemed like magic.

When I was in college, I learned to make websites through careful, painstaking effort.

When I was a young professional, I used those skills and others to make websites for hospitals and summer camps and conferences.

Then I learned software development and practiced the slow, methodical process of writing and debugging software.

Now, I get to make beautiful things by speaking, guiding, and directing a system which is capable of handling the drudgery while I think about how to make the system wonderful and functional and beautiful.

It was, for me, never about the code. It was always about making something useful for myself and others. And that has never been easier.

"Wait 6 months" has been the call for 3-4 years now. You can't eulogize a profession that hasn't been killed, that's just mean.

This perspective was mine 6 months ago. And god damn, I do miss the feeling of crafting something truly beautiful in code sometimes. But then, as I've been pushed into this new world we're living in, I've come to realize a couple things:

Nothing I've ever built has lasted more than a few years. Either the company went under, or I left and someone else showed up and rewrote it to suit their ideals. Most of us are doing sand art. The tide comes in and its gone.

Code in and of itself should never have been the goal. I realized that I was thinking of the things I build and the problems I selected to work on from the angle of code quality nearly always. Code quality is important! But so is solving actual problems with it. I personally realized that I was motivated more by the shape of the code I was writing than the actual problems it was written to solve.

Basically the entire way I think about things has changed now. I'm building systems to build systems. Thats really fun. Do I sometimes miss the feeling of looking at a piece of code and feeling a sense of satisfaction of how well made it is? Sure. That era of software is done now sadly. We've exited the craftsman era and entered into the Ikea era of software development.

> Now is the time to mourn the passing of our craft.

Your craft is not my craft.

It's entirely possible that, as of now, writing JavaScript and Java frontends (what the author does) can largely be automated with LLMs. I don't know who the author is writing to, but I do not mistake the audience to be "programmers" in general...

If you are making something that exists, or something that is very similar to something that exists, odds are that an LLM can be made to generate code which approximates that thing. The LLM encoding is lossy. How will you adjust the output to recover the loss? What process will you go through mentally to bridge the gap? When does the gap appear? How do you recognize it? In the absolute best case you are given a highly visible error. Perhaps you've even shipped it, and need to provide context about the platform and circumstances to further elucidate. Better hope that platform and circumstance is old-hat.

"They can write code better than you or I can"

They can not. They can make some average code. On Friday one suggested an NSI installer script that would never bundle some needed files in the actual installer. I can only imagine that a lot of people have made the same mistake (used CopyFiles instead of File) and posted that mistake on the internet. The true disaster of that being that then testing out that installer on the developer's PC, where that CopyFiles may well work fine since the needed files happen to be sitting on that PC, would then lead on to think it was some weird bug that only failed on the end user's PC. I bet a lot of people posted it with comments like "this worked fine when I tried it," and here we are a decade later feeding that to an LLM.

These tools can write average code. That's what they've mostly been fed; that's what they're aiming for when they do their number crunching. The more specifically one prompts, I expect, then the more acceptable that average code will be. In some cases, average appears to be shockingly bad (actually, based on a couple of decades' experience in the game, average is generally pretty bad - I surely must have been churning out some average, bad code twenty years ago). If I want better than average, I'm going to have to do it myself.

I feel like most of the anxiety around LLMs is because (in the USA at least) our social safety net sucks.

I'd probably have way more fun debating LLMs if it wasn't tied to my ability to pay rent, have healthcare, or feel like a valued person contributing something to society. If we had universal healthcare and a federal job guarantee it would probably calm things down.

"They can write code better than you or I can, and if you don’t believe me, wait six months." They've been saying that for years. Stop believing it.

I dont get the hype.. And I dont think we will reach peak AI coding performance any time soon.

Yes, watching an LLM spit out lots of code is for sure mesmerizing. Small tasks usually work ok, code kinda compiles, so for some scenarios it can work out.. BUT anyone serious about software development can see how piece of CRAP the code is.

LLMs are great tools overall, great to bounce ideas, great to get shit done. If you have a side project and no time, awesome.. If your boss/company has a shitty culture and you just want to get the task done, great. Got a mundane coding task, hate coding, or your code wont run in a critical environment? please, LLM that shit over 9000..

Remember though, an LLM is just a predictor, a noisy, glorified text predictor. Only when AI reaches a point of not optimizing for short term gains and has built-in long term memory architecture (similar to humans) AND can produce some linux kernel level code and size, then we can talk..

Programming brings me joy in two different ways.

1. Crafting something beautiful. Figuring out correct abstractions and mapping them naturally to language constructs. Nailing just the right amount of flexibility, scalability and robustness. Writing self-explanatory, idiomatic code that is a pleasure to read. It’s an art.

2. Building useful things. Creating programs that are useful to myself and to others, and watching them bring value to the world. It’s engineering.

These things have utility but they are also enjoyable onto themselves. As best I can tell, your emotional response to coding agents depends on how much you care about these two things.

AI has taken away the joy of crafting beautiful things, and has amplified the joy of building things by more than 10x. Safe bet: It will get to 100x this year.

I am very happy with this tradeoff. Over the years I grew to value building things much more highly. 20yo me would’ve been devastated.

> I didn’t ask for the role of a programmer to be reduced to that of a glorified TSA agent, reviewing code to make sure the AI didn’t smuggle something dangerous into production.

This may be the perspective of some programmers. It doesn't seem to be shared by the majority of software engineers I know and read and listen to.

" They can write code better than you or I can, and if you don’t believe me, wait six months."

It's ALWAYS wait 6 months, or wait for the next generation. Or "oh that model that we told you to use is old now, use this new one instead. Oh that doesn't work? Well that's old now, use this one". Always. 6 months ago it was wait 6 months. 12 months ago it was wait 6 months. 18 months ago it was wait 6 months. Now it's wait 6 months. 6 months from now it'll be wait 6 months.

I often venerate antiques and ancient things by thinking about how they were made. You can look at a 1000-year-old castle and think: This incredible thing was built with mules and craftsmen. Or look at a gorgeous, still-ticking 100-year-old watch and think: This was hand-assembled by an artist. Soon I'll look at something like the pre-2023 Linux kernel or Firefox and think: This was written entirely by people.

From a blog post last month by the same author:

> Today, I would say that about 90% of my code is authored by Claude Code. The rest of the time, I’m mostly touching up its work or doing routine tasks that it’s slow at, like refactoring or renaming.

> I see a lot of my fellow developers burying their heads in the sand, refusing to acknowledge the truth in front of their eyes, and it breaks my heart because a lot of us are scared, confused, or uncertain, and not enough of us are talking honestly about it. Maybe it’s because the initial tribal battle lines have clouded everybody’s judgment, or maybe it’s because we inhabit different worlds where the technology is either better or worse (I still don’t think LLMs are great at UI for example), but there’s just a lot of patently unhelpful discourse out there, and I’m tired of it.

https://nolanlawson.com/2026/01/24/ai-tribalism/

If you're responding to this with angry anti-AI rants (or wild AI hype), might want to go read that post.

> They can write code better than you or I can, and if you don’t believe me, wait six months.

You can use AI to write all your code, but if you want to be a programmer and can't see that the code is pretty mid then you should work on improving your own programming skills.

People have been saying the 6 month thing for years now, and while I do see it improving in breadth, quality/depth still appears to be plateauing.

It's okay if you don't want to be a programmer though, you can be a manager and let AI do an okay job at being your programmer. You better be driven to be a good at manager though. If you're not... then AI can do an okay job of replacing you there too.

I get the grief about AI, but I don't share it.

After ten years of professional coding, LLMs have made my work more fun. Not easier in the sense of being less demanding, but more engaging. I am involved in more decisions, deeper reviews, broader systems, and tighter feedback loops than before. The cognitive load did not disappear. It shifted.

My habits have changed. I stopped grinding algorithm puzzles because they started to feel like practicing celestial navigation in the age of GPS. It is a beautiful skill, but the world has moved on. The fastest path to a solution has always been to absorb existing knowledge. The difference now is that the knowledge base is interactive. It answers back and adapts to my confusion.

Syntax was never the job. Modeling reality was. When generation is free, judgment becomes priceless.

We have lost something, of course. There is less friction now, which means we lose the suffering we often mistook for depth. But I would rather trade that suffering for time spent on design, tradeoffs, and problems that used to be out of reach.

This doesn't feel like a funeral. It feels like the moment we traded a sextant for a GPS. The ocean is just as dangerous and just as vast, but now we can look up at the stars for wonder, rather than just for coordinates.

If AI is good enough that juniors wielding it outproduce seniors, then the juniors are just... overhead. The company would cut them out and let AI report to a handful of senior architects who actually understand what's being built. You don't pay humans to be a slow proxy for a better tool.

If the tools get good enough to not need senior oversight, they're good enough to not need junior intermediaries either. The "juniors with jetpacks outpacing seniors" future is unrealistic and unstable—it either collapses into "AI + a few senior architects" or "AI isn't actually that reliable yet."

I’m in my 40 something and it’s game over for my career. The grey in my hair makes it so that I never get past the first round. The history on my resume makes it so I’m lucky to get a round. The GPT’s and Claude have fundamentally changed how I view work and frankly, I’m over it.

I’m in consulting now and it’s all the same crap. Enterprises want to “unleash AI” so they can fire people. Maximize profits. My nephews who are just starting their careers are blindly using these tools and accepting the PR if it builds. Not if it’s correct.

I’m in awe of what it can do but I also am not impressed with the quality of how it does it.

I’m fortunate to not have any debt so I can float until the world either wises up or the winds of change push me in a new direction.

I liked the satisfaction of building something “right” that was also “useful”. The current state of Opus and Codex can only pretend to do the latter.

I'm surprised so many people are only waking up to this now. It should have been obvious as soon as ChatGPT came out that even with only incremental improvements, LLMs would kill programming as we knew it. And the fact that these utterances, however performative, from developers expressing grief or existential despair have become commonplace tells me as much about the power of these systems than whatever demo Anthropic or OpenAI has cooked up.

I would also point out that the author, and many AI enthusiasts, still make certain optimistic assumptions about the future role of "developer," insisting that the nature of the work will change, but that it will somehow, in large measure, remain. I doubt that. I could easily envision a future where the bulk of software development becomes something akin to googling--just typing the keywords you think are relevant until the black box gives you what you want. And we don't pay people to google, or at least, we don't pay them very much.

People have to stop talking like LLMs solved programming.

If you're someone with a background in Computer Science, you should know that we have formal languages for a reason, and that natural language is not as precise as a programming language.

But anyway we're peek AI hype, hitting the top on HN is worth more than a reasonable take, reasonableness doesn't sell after all.

So here we're seeing yet another text about how the world of software was solved by AI and being a developer is an artifact of the past.

I do not mourn typing in code.

But I am still quite annoyed at the slopful nature of the code that is produced when you're not constantly nagging it to do better

We've RLed it to produce code that works by hook or by crook, putting infinity fallback paths and type casts everywhere rather than checking what the semantics should be.

Sadly I don't know how we RL taste.

I was very impressed by AI-generated CSS when I didn't know CSS.

Then I learned CSS.

Now I am not as impressed by AI-generated CSS.

I hope that "our craft" which now produces, largely, vulnerable buggy bloatware actually dies.

Perhaps people or machines will finally figure out how to make software which actually works without a need to weekly patching

> They can write code better than you or I can

Speak for yourself. They produce shit code and have terrible judgment. Otherwise we wouldn't need to babysit them so much.

“We’ll miss creating something we feel proud of”

I still feel proud when skillfully guiding a set of AI agents to build from my imagination. Especially when it was out of my reach just 6-months ago.

I’m a 49 year old veteran who started at just 10 years old and have continued to find pure passion in it.

Some code is worth transcribing by hand — an ancient practice in writing, art and music.[0] Some isn't even worth looking at.

I find myself, ironically, spending more time typing out great code by hand now. Maybe some energy previously consumed by tedium has been freed up, or maybe the wacky machines brought a bit of the whimsy back into the process for me.

[0] And in programming, for the readers of Zed Shaw's books :)

I don't mourn coding for itself, since I've always kinda disliked that side of my work (numerical software, largely).

What I do mourn is the reliability. We're in this weird limbo where it's like rolling a die for every piece of work. If it comes up 1-5, I would have been better off implementing it myself. If it comes up 6, it'll get it done orders of magnitude faster than doing it by hand. Since the overall speedup is worthwhile, I have to try it every time, even if most of the time it fails. And of course it's a moving target, so I have to keep trying the things that failed yesterday because today's models are more capable.

> We’ll miss the feeling of holding code in our hands

I agree, I started feeling this a few months ago, where I was only writing the architecture and abstractions and letting AI fill in the gaps. It seems in the next few months it could probably do more than that. But is it so bad, I agree that I can't really mold an entire pot by my hand any more. But, if you ask AI to do it, it will create a pot with cracks in it and it would be your job to either plaster it of fill gold in those cracks.

I feel coding is going to be similar to kintsugi after this is all over

> We’ll miss the sleepless wrangling of some odd bug that eventually relents to the debugger at 2 AM.

I'll miss it not because the activity becomes obsolete, but because it's much more interesting than sitting till 2am trying to convince LLM to find and fix the bug for me.

We'll still be sitting till 2am.

> They can write code better than you or I can, and if you don’t believe me, wait six months.

I've been hearing this for the last two years. And yet, LLMs, given abstract description of the problem, still write worse code than I do.

Or did you mean type code? Because in that case, yes, I'd agree. They type better.

Like other tech disrupted crafts before this, think furniture making or farming, that's how it goes. From hand-made craft, to mass production factories (last couple of decades) to fully automated production.

The craft was dying long before LLMs. Started in dotcom, ZIRP added some beatings, then LLMs are finishing the job.

This is fine, because like in furniture making, the true craftsmen will be even more valuable (overseeing farm automation, high end handmade furniture, small organic farms), and the factory worker masses (ZIRP enabled tech workers) will move on to more fulfulling work.

Dang right we do.

And the problem isn't even the Junior Zoomer devs running circles around seniors. It's the CTO or Engineering VP himself disappearing for a few months and single-handedly consolidating a handful of products into a full rewrite for the company, excluding most of their engineering team from the process, and then laying them off after.

The problem is the CEO pretending to be an engineer and thinking they know better because they can write English prompts and spit out a hideous prototype.

The problem is Product Owners using LLMs to "write code" while their engineering team does zero human review before merging it, because their AI tooling was made solely responsible for code quality. If something's broken, just prompt a sloppy fix full of hidden performance and security bugs that the automated code review step missed.

If you think this is hyperbole, I was recently laid off from a company that did exactly the above.

Then in 2027, it will be product owners replacing the entire engineering team, including the CTO, because they made their system too reliable to justify their employment, while the "thinkers" get to build the product, engineers be damned.

People with real skills they acquired over a lifetime are no longer shaping business. Reckless efficiency towards being average will rule the day.

I fall in the demographic discussed in the article but I’ve approached this with as much pragmatism as I can muster. I view this as a tool to help improve me as a developer. Sure there will be those of us who do not stay ahead (is that even possible?) of the curve and get swallowed up but technology has had this affect on many careers in the past. They just change into something different and sometimes better. It’s about being willing to change with it.

I agree with the nolanlawson's sentiment. What's interesting is many of the opposing statements here seem to be less interested in the actual code, and more interested in the final state. Both are valid, but one is going away due to technological advancements. That is the mourning.

There are some of us who enjoyed the code as a thing to explore. Others here don't seem to like that as much.

I've been hearing "the LLM can write better code than a human, and if you don't believe me, wait six months" for years now. Such predictions haven't been true before and I don't believe they are true now.

I can relate coz once upon a time I enjoyed reading poetic code - which I later found to be impossible to modify or extend.

> Eventually your boss will start asking why you’re getting paid twice your zoomer colleagues’ salary to produce a tenth of the code.

And then I couldn’t relate because no one ever paid me for lines of code. And the hardest programming I ever did was when it took me 2 days to write 2 lines of C code, which did solve a big problem.

I abhorred the LOC success metric because we had to clean up after those who dumped their code diarrhea to fool those who thought every line of code is added value. Not to mention valuing LOC strictly makes you a junior programmer.

E.g. you have to know more to do the following in 2 lines (yes, you can):

‘’’ t = a; a = b; b = a; ‘’’

According to LOC missionaries these 3 lines are more expensive to write and shows that you’re a better programmer than the XOR swap. But it’s actually more expensive to run it than the XOR swap, and it’s more expensive to hire the person who can write the XOR swap. (Not endorsing clever/cryptic code, just making a point about LOC)

So, if the LOC missionaries are out of a job because of LLMs, I will probably celebrate.

I think you need to lighten up and adapt the way programmers have adapted since the 50s. There’s been a lot of pomposity over the past ten years around “craft”, you’d think we’re all precious Renaissance artists pining for recognition over our heavenly crafted code rather than engineers. I’m in my late forties also I’m case that matters at all, I don’t think it does. If anything that gives me an advantage to spot and quickly filter out the 1/10 insanely bad suggestions that AI pukes out and know when it’s giving me decent code. Who knows where it will lead but stubbornly refusing to move won’t help. Please take this in a positive way also, rather than a dig. Be positive. Also think of all the job opportunities you will have rewriting all the crap code that gets punted out over the next 10 years. You’re going to be rich. Seriously tho, if you are in your forties you have the advantage of leveraging AI yourself but also knowing when it is bull shitting you.

> So as a senior, you could abstain. But then your junior colleagues will eventually code circles around you, because they’re wearing bazooka-powered jetpacks and you’re still riding around on a fixie bike. Eventually your boss will start asking why you’re getting paid twice your zoomer colleagues’ salary to produce a tenth of the code.

I might be mistaken, but I bet they said the same when Visual Basic came out.

Dunno, LLMs writing code still feels like they memorized a bunch of open source code and vomited them out in worse condition.

It's not that impressive that Claude wrote a C compiler when GitHub has the code to a bunch of C compilers (some SOTA) just sitting there.

I'm using an LLM to write a compiler in my spare time (for fun) for a "new" language. It feels more like a magical search engine than coding assistant. It's great for bouncing ideas from, for searching the internet without the clutter of SEO optimized sites and ads, it's definitely been useful, just not that useful for code.

Like, I have used some generated code in a very low stakes project (my own Quickshell components) and while it kind of worked, eventually I refactored it myself into 1/3 of the lines it produced and had to squash some bugs.

It's probably good enough for the people who were gluing React components together but it still isn't on the level where I'd put any code it produces into production anywhere I care about.

There's a commercial building under construction next to my office. I look down on the construction site, and those strapping young men are digging with their big excavators they've been using for years and taking away the dirt with truck and trailer.

Why use a spade? Even those construction workers use the right sized tools. They ain't stupid.

It isn't about the tools or using them, it's about the scale. The scale of impact is immense and we're not ready to handle it in a mutitude of areas because of all the areas technology touches. Millions of jobs erased with no clear replacement? Value of creative work diminshed leading to more opportunities erased? Scale of 'bad' actors abusing the tools and impacting a whole bunch of spheres from information dispersal to creative industries etc. Not even getting into environmental and land-use impacts to spaces with data centers and towns etc (again, it's the scale that gets ya). And for what? Removing a huge chunk of human activity & expression, for what?

This makes me think about the craftsmen whose careers vanished or transformed through the ages due to industries, machines etc. They did not have online voices to write 1000's of blogs everyday. Nor did they have people who can read their woes online.

I get where this is coming from. But at the same, AI/LLMs are such an exciting development. As in "maybe I was wrong and the singularity wasn't bullshit". If nothing else, it's an interesting transition to live through.

Write a blog post promoting inevitability of AI in software development while acknowledging feelings of experienced software engineers.

the robot does to programming what the wall street did to thievery.

the police knew what's up. the law knew what's up. people knew what's up. you could be vigilant and prepared and reimbursed if the police managed to catch the thief. then came wall street and thieves (mostly) mourned their craft.

the law has no idea. the police has no idea. the people have no idea. it's all fraud. no honor, no skill, no chance to reimburse those who are defrauded.

jobs? demand creates jobs. doing things the right way creates jobs. not greedy, fuzzy business logic. that killed more jobs and SMEs than it created. killed more agency than it created opportunities. we know, it's just logic. but the law, the police, engineers looked the other way and worse, supported wall street. it's fun, I'm sure.

but don't think the world doesn't know what role programmers played. mourn all you want.

programmers had the chance to fight shady corporations and bad business practices and the darkest of dark patterns but they shat on teens and their attention and blamed the individual and worked for facebook and wall street and played key roles in expanding fraud, bad business practices and implemented more and more dark patterns and ways to hijack parents and teens minds, and to stress parents and kill agency and attention and care and vigilance.

they had a chance. now they mourn their craft.

I'm exaggerating. Relax. Y'all be fine. Others won't.

It's always 'eventually', 'soon', 'by the latter half of this year' with AI. Meanwhile I am both bemused and horrified as I watch a bunch of idiots desperate ly trying to find the magical 10x trick with AI unleash more and more mediocre code full of security vulnerabilities, running on agents with root permissions, that I somehow have to 'secure' for the company :D At least I have job security! Up until we get hacked sideways at least.

I on the other hand await the coming of the Butlerian Jihad.

> wait six months

from other sources: 6-12 months, by end of 2025, ChatGPT 7.

It's a concern trolling and astroturfing at it best.

One camp of fellow coders who are saying how their productivity grew 100x but we are all doomed, another camp of AI enthusiasts who got ability to deliver products and truly believe in their newly acquired superiority.

It's all either true or false, but if in six months it becomes true, we'll know it, each one of us.

However, if it's all BS and in six months there will be Windows 95 written by LLM but real code still requires organic intelligence, there won't be any accountability, and that's sad.

I've noticed a steep decline in software quality as a consumer since Covid, but especially since ChatGPT came out.

That hasn't changed. Whether it's apps that hog too much memory, games that use too much storage, unresponsiveness, or just plain cryptic error messages, everything feels more fraile than it used to be.

Perhaps its the growing pains of LLMs, a horde of junior programmers pushing stuff to production that the seniors were too "old fashioned" to notice.

Only time will tell.

My ability to ask questions & hone in on good answers is far better than it ever was. My ability to change course & iterate is far faster than it ever has been. I'm making far more informed decisions, far more able to make forays and see how things turn out, with low cost.

I could not be having a better time.

I liked coding! It was fun! But I mourned because I felt like I would never get out 1% of the ideas in my head. I was too slow, and working on shit in my free time just takes so much, is so hard, when there's so little fruitful reward at the end of a weekend.

But I can make incredible systems so fast. This is the craft I wanted to be doing. I feel incredibly relieved, feel such enormous weigh lifted, that maybe perhaps some of my little Inland Empire that lives purely in my head might perhaps make it's way to the rest of the world, possibly.

Huge respect for all the sadness and mourning. Yes too to that. But I cannot begin to state how burdened and sad I felt, so unable to get the work done, and it's a total flip, with incredible raw excitement and possibility before me.

That said, software used to reward such obsessive deep following pursuit, such leaning into problems. And I am very worried, long term, what happens to the incredible culture of incredible people working really hard together to build amazing systems.

The acceleration of AI has thrown into sharp relief that we have long lumped all sorts of highly distinct practices under this giant umbrella called "coding". I use CC extensively, and yet I still find myself constantly editing by hand. Turns out CC is really bad at writing kubernetes operators. I'd bet it's equally bad at things like database engines or most cutting edge systems design problems. Maybe it will get better at these specific things with time, but it seems like there will always be a cutting edge that requires plenty of human thought to get right. But if you're doing something that's basically already been done thousands of times in slightly different ways, CC will totally do it with 95% reliability. I'm ok with that.

It's also important to step back and realize that it goes way beyond coding. Coding is just the deepest tooth of the jagged frontier. In 3 years there will be blog posts lamenting the "death of law firms" and the "death of telemedicine". Maybe in 10 years it will be the death of everything. We're all in the same boat, and this boat is taking us to a world where everyone is more empowered, not less. But still, there will be that cutting edge in any field that will require real ingenuity to push forward.

    We’ll miss creating something we feel proud of, something true and right and good. We’ll miss the satisfaction of the artist’s signature at the bottom of the oil painting, the GitHub repo saying “I made this.”

I definitely feel this some days. It used to be that I would nitpick code to death to get it just so, proud of the artistic decisions I made. Not only was it functional, but it was beautiful, crisp, elegant, clever.

Just the right abstractions. Complicated problems reduced to perfectly legible structures, easy to read for newcomers, easy to extend when new problems arise.

Extracting that from an LLM just doesn't produce the same feeling, even when it produces the same results. And they don't reliably produce the same results yet even, just poor imitations, however functional.

On the other hand, when they work, I am empowered to produce code I would have never thought of. I'm empowered to bounce an idea off an oracle that knows all the answers, and can tailor its responses to my exact use case. I'm coming around to finding the joy in that. I have to.

To the people who are against AI programming, honest question: why do you not program in assembly? Can you really say "you" "programmed" anything at all if a compiler wrote your binaries?

This is a 100% honest question. Because whatever your justification to this is, it can probably be used for AI programmers using temperature 0.0 as well, just one abstraction level higher.

I'm 100% honestly looking forward to finding a single justification that would not fit both scenarios.

If you want to build a house you still need plans. Would you rather cut boards by hand or have a power saw. Would you rather pound nails, pilot hole with a bit and brace and put in flat head screws... or would you want a nail gun and an impact driver.

And you still need plans.

Can you write a plan for a sturdy house, verify that it meets the plan that your nails went all the way in and in the right places?

You sure can.

Your product person, your directors, your clients might be able to do the same thing, it might look like a house but its a fire hazard, or in the case of most LLM generated code a security one.

The problem is that we moved to scrum and agile, where your requirements are pantomime and postit notes if your lucky, interpretive dance if you arent. Your job is figuring out how to turn that into something... and a big part of what YOU as an engineer do is tell other people "no thats dumb" without hurting their feelings.

IF AI coding is going to be successful then some things need to change: Requirements need to make a come back. GOOD UI needs to make a comeback (your dark pattern around cancelation, is now going to be at odds with an agent). Your hide the content behind a login or a pay wall wont work any more because again, end users have access too... the open web is back and by force. If a person can get in, we have code that can get in now.

There is a LOT of work that needs to get done, more than ever, stop looking back and start looking forward, because once you get past the hate and the hype there is a ton of potential to right some of the ill's of the last 20 years of tech.

> because they’re wearing bazooka-powered jetpacks and you’re still riding around on a fixie bike

Sure, maybe it takes me a little while to ride across town on my bike, but I can reliably get there and I understand every aspect of the road to my destination. The bazooka-powered jetpack might get me there in seconds, but it also might fly me across state lines, or to Antarctica, or the moon first, belching out clouds of toxic gas along the way.

I am at the tail-end of my career and I do not feel that option to ignore AI. Thankfully I am soberly fascinated with these developments. Also, I often find it is easier to articulate my coding desire with actual code, than with a descriptive prompt. And I have definitely seen AI fail in difficult scenarios. I am often needed.

I'm that 40 year old now. Been writing code since grade 5. Loved it so much I got a PhD, was an academic, then moved into industry.

I don't mourn or miss anything. No more then the previous generation mourned going from assembly to high level languages.

The reason why programming is so amazing is getting things done. Seeing my ideas have impact.

What's happening is that I'm getting much much faster and better at writing code. And my hands feel better because I don't type the code in anymore.

Things that were a huge pain before are nothing now.

I didn't need to stay up at night writing code. I can think. Plan. Execute at a scale that was impossible before. Alone I'm already delivering things that were on the roadmap for engineering months worth of effort.

I can think about abstractions, architecture, math, organizational constraints, product. Not about what some lame compiler thinks about my code.

And if someone that's far junior to me can do my job. Good. Then we've empowered them and I've fallen behind. But that's not at all the case. The principals and faculty who are on the ball are astronomically more productive than juniors.

This entire panic is a mass-hysteria event. The hallucination that "an LLM can do software engineering better than a 10x engineer" is only possible because there are so few 10xers left in the business. 99% either retired or are otherwise not working at the moment.

The "difficult", "opinionated", "overpaid" maniacs are virtually all gone. That's why such a reckless and delusional idea like "we'll just have agents plan, coordinate, and build complete applications and systems" is able to propagate.

The adults were escorted out of the building. Managements' hatred of real craftspeople is manifesting in the most delusional way yet. And this time, they're actually going to destroy their businesses.

I'm here for it. They're begging to get their market share eaten for breakfast.

I suspect my comment will not be well received, however I notice in myself that I've passed the event horizon of being a believer and am past the honeymoon period and I'm beginning to think about engineering

My headspace is now firmly in "great, I'm beginning to understand the properties and affordances of this new medium, how do I maximise my value from it", hopefully there's more than a few people who share this perspective, I'd love to talk with you about the challenges you experience, I know I have mine, maybe we have answers to each others problems :)

I assume that the current set of properties can change, however it seems like some things are going to be easier than others, for example multi modal reasoning still seems to be a challenge and I'm trying to work out if that's just hard to solve and will take a while or if we're not far from a good solution

I thought I'd miss all the typing and syntax, but I really don't. Everyone has their own relationship with coding, but for me, I get satisfaction out of the end product and putting it in front of someone. To the extend that I cared about the code, it mainly had to do with how much it allowed the end product to shine.

That's why I'll only read source code written until 2024.

> Someday years from now we will look back on the era when we were the last generation to code by hand. We’ll laugh and explain to our grandkids how silly it was that we typed out JavaScript syntax with our fingers. But secretly we’ll miss it.

Why will I miss it? I will be coding my small scripts and tools and hobby projects by hand because there is no deadline attached to them. Hell, I will also tag them as "bespoke hand-crafted free range artisanal" code. There will be a whole hipster category of code that is written as such. And people will enjoy it as they enjoy vinyl records now. Many things would have changed by then but my heart will still be a programmer's heart.

An architect knows the difference between this window and that, this door and another, a partition wall and a firewall, a tile roof and a metal one. The architect understands how to orient them and work the materials together.

But the architect does not have to make the windows, or the doors, or the walls…or the roof to create the house.

There are lots of things that autonomous agents are not so good at. Most of the things it took away from me are things that were holding me back in the first place.

1. it isn't that bad

2. the tools still need a lot of direction, i still fight claude with opus to do basic things and the best experiences are when i provide very specific prompts

3. being idealistic on a capitalist system where you have to pay your bills every month is something i could do when my parents paid my bills

These apocalyptic posts about how everything is shit really don't match my reality at all. I use these tools every day to be more productive and improve my code but they are nowhere close to doing my actual job, that is figuring out WHAT to do. How to do it is mostly irrelevant, as once i get to that point i already know what needs to be done and it doesn't matter if it is me or Opus producing the code.

Think there is some reversion to wider mean at play here too. The majority of jobs out there are not about whether the employee feels fulfilled. And commoditisations of code is likely to trigger same here.

The majority of the code currently running in production for my company was written 5+ years ago. This was all "hand-written" and much lower quality than the AI generated code that I am generating and deploying these days.

Yet I feel much more connected with my old code. I really enjoyed actually writing all that code even though it wasn't the best.

If AI tools had existing 5 years ago when I first started working on this codebase, obviously the code quality would've been much higher. However, I feel like I really loved writing my old code and if given the same opportunity to start over, I would want to rewrite this code myself all over again.

I feel like a lot of comments here are missing the point. I think the article does a fairly good job neither venerating nor demonizing AI, but instead just presenting it as the reality of the situation, and that reality means that the craft of programming and engineering is fundamentally different than it was just a few years ago.

As an (ex-)programmer in his late 40s, I couldn't agree more. I'm someone who can be detail-oriented (but, I think also with a mind toward practicality) to the point of obsession, and I think this trait served me extremely well for nearly 25 years in my profession. I no longer think that is the case. And I think this is true for a lot of developers - they liked to stress and obsess over the details of "authorship", but now that programming is veering much more towards "editor", they just don't find the day-to-day work nearly as satisfying. And, at least for me, I believe this while not thinking the change to using generative AI is "bad", but just that it's changed the fundamentals of the profession, and that when something dies it's fine to mourn it.

If anything, I'm extremely lucky that my timing was such that I was able to do good work in a relatively lucrative career where my natural talents were an asset for nearly a quarter of a century. I don't feel that is currently the case regarding programming, so I'm fortunate enough to be able to leave the profession and go into violin making, where my obsession with detail and craft is again a huge asset.

An excellent read.

I mourn and i celebrate. I can't think of another craft for which the implicit goal was to make itself obsolete, but we've just about done it!

and that is something to celebrate

I didn't come in IT for money - back in the days it wasn't as well paid as today - nevertheless if this craft was very poorly paid I probably wouldn't choose this profession either. And I assume many people here wouldn't as well unless you are already semi-retired or debt free.

I mourn a little bit that in 20 years possibly 50% of software jobs will get axed or unless you are elite/celebrity dev salary will stagnate. I mourn that in the future upward mobility and moving up into upper middle class will be harder without trying to be entrepreneur.

I'm probably a minority, but I've never loved dealing with syntax. The code itself always felt like a hindrance to me that reminded me that my brain was slowed down by my fingers. I get it though, it was tactical and it completed the loop. It was essential for learning I felt like despite eventually getting to a point where it slows you down the more senior you get.

AI has a ways to go before it's senior level if it ever reaches that level, but I do feel bad for juniors that survive this who never will have the opportunity to sculpt code by hand.

I'll believe it when I start seeing examples of good and useful software being created with LLMs or some increase in software quality. So far it's just AI doom posting, hype bloggers that haven't shipped anything, anecdotes without evidence, increase in CVEs, increase in outages, and degraded software quality.

It was pretty clear to me after interacting with the first popular ChatGPT version around end of 2022 that all knowledge jobs will be replaced sooner or later. I don’t think coding is somehow special. What we have right now is an intermediate stage where “taste” still matters, but this won’t last forever.

I also believe that when all knowledge jobs are replaced, something fundamental needs to change in society. Trying to anticipate and prepare for that right now is premature.

I feel like we are long into the twilight of mini blogs and personal sites. Its like people trying to protect automotive jobs, the vas majority were already lost

Perhaps Im a cynic but I don't know

> They can write code better than you or I can, and if you don’t believe me, wait six months.

I feel like I've been hearing this exact sentence for 2 years now...

I wonder whether, in the end, it was simply poor accessibility that made programmers special, and whether it is that what some of them are missing. Being special by "talking" a special language their customers can't comprehend.

Sure, they are still needed for debugging and for sneering at all those juniors and non-programmers who will finally be able to materialise their fantasies, but there is no way back anymore, and like riding horses, you can still do it while owning a car.

> If you would like to grieve, I invite you to grieve with me.

I think we should move past this quickly. Coding itself is fun but is also labour , building something is the what is rewarding.

So funny because "you’re still riding around on a fixie bike." that's literally what I do. I also take a plane to fly.

In that analogy that would be picking the right library/framework/service vs. vibe coding it. I do NOT need to write all the code to provide value.

Knowing which tool to use for the task is precisely what requires honing a skill and requires judgement. It's not about how many kilometers you manage to cover.

Some years ago I was at the Burger King near the cable car turntable at Powell and Market St in San Francisco. Some of the homeless people were talking about the days when they'd been printers. Press operators or Linotype operators. Jobs that had been secure for a century were just - gone.

That's the future for maybe half of programmers.

Remember, it's only been three years since ChatGPT. This is just getting started.

> and parrot it back

I don't like this particular phrase, as it suggests that LLMs are just replicating code. As far as I understand, LLMs can also create new code and algorithms (some better than others). When we think of them as copy-paste-maschines, we judge their capabilities unfairly and also underestimate their capabilities.

I understand the sentiment and I've been involved in software engineering in various roles for the last 25+ years. The thing that gives me hope is that never once in that time has the problem ever been that we didn't have more work to do.

It's not like all of a sudden I'm working 2-3 hours a day. I'm just getting a lot more done.

Two years ago I decided to give up my career as an industry researcher to pursue a tenure-track professor position at a community college. One of the reasons I changed careers is because I felt frustrated with how research at my company changed from being more self-directed and driven by longer-term goals to being directed by upper management with demands for more immediate productization.

I feel generative AI is being imposed onto society. While it is a time-saving tool for many applications, I also think there are many domains where generative AI needs to be evaluated much more cautiously. However, there seems to be relentless pressure to “move fast and break things,” to adopt technology due to its initial labor-saving benefits without fully evaluating its drawbacks. That’s why I feel generative AI is an imposition.

I also resent the power and control that Big Tech has over society and politics, especially in America where I live. I remember when Google was about indexing the Web, and I first used Facebook when it was a social networking site for college students. These companies became successful because they provided useful services to people. Unfortunately, once these companies gained our trust and became immensely wealthy, they started exploiting their wealth and power. I will never forget how so many Big Tech leaders sat at Trump’s second inauguration, some of whom got better seats than Trump’s own wife and children. I highly resent OpenAI’s cornering of the raw wafer market and the subsequent exorbitant hikes in RAM and SSD prices.

Honestly, I have less of an issue with large language models themselves and more of an issue with how a tiny handful of powerful people get to dictate the terms and conditions of computing for society. I’m a kid who grew up during the personal computing revolution, when computation became available to the general public. I fell for the “computers for the rest of us,” “information at your fingertips” lines. I wanted to make a difference in the world through computing, which is why I pursued a research career and why I teach computer science.

I’ve also sat and watched research industry-wide becoming increasingly driven by short-term business goals rather than by long-term visions driven by the researchers themselves. I’ve seen how “publish-and-perish” became the norm in academia, and I also saw DOGE’s ruthless cuts in research funding. I’ve seen how Big Tech won the hearts and minds of people, only for it to leverage its newfound power and wealth to exploit the very people who made Big Tech powerful and wealthy.

The tech industry has changed, and not for the better. This is what I mourn.

You know who else mourned the loss of craft? People that don't like PHP and Wordpress because they lower the barrier to entry to creating useful stuff while also leaving around a fair amount of cruft and problems that the people that use them don't understand how to manage.

Like iambateman said: for me it was never about code. Code was a means to an ends and it didn't stop at code. I'm the kind of software engineer that learned frontends, systems, databases, ETLs, etc -- whatever it was that was that was demanded of me to produce something useful I learned and did it. We're now calling that a "product engineer". The "craft" for me was in creating useful things that were reliable and efficient, not particularly how I styled lines, braces, and brackets. I still do that in the age of AI.

All of this emotional spillage feels for not. The industry is changing as it always has. The only constant I've ever experienced in this industry is change. I realized long ago that when the day comes that I am no longer comfortable with change then that is my best signal that this industry is no longer for me.

One other helpful frame: I consider LLMs simply to be very flexible high-level 'language' Compilers. We've moved up the Abstraction Chain ever since we invented FORTRAN and COBOL (and LISP) instead of using assembly language.

We're 'simply' moving up the abstraction hierarchy again. Good!

The death of a means to an end is the birth of an end itself.

When cameras became mainstream, realism in painting went out of fashion, but this was liberating in a way as it made room for many other visual art styles like Impressionism. The future of programming/computing is going to be interesting.

We have CNC machines, and we still have sculptors.

Mechanising the production of code is good thing. And crafting code as art is a good thing. It is sign of a wider trend that we need to look at these things like adversaries.

I look forward to the code-as-art countermovement. It's gonna be quite something.

Great post. Super sad state of affairs but we move on and learn new things. Programming was always a tool and now the tool has changed from something that required skill and understanding to complaining to a neural net. Just have to focus on the problem being solved more.

Coding is an abstraction. Your CPU knows nothing of type safety, bloom filters, dependencies, or code reuse.

Mourning the passing of one form of abstraction for another is understandable, but somewhat akin to bemoaning the passing of punch card programming. Sure, why not.

The thing he has spent his whole career doing unto others he finally did into himself

This is probably a good representative take for the view of people who assigned value to the coding part of building software. It's worth reading and saving, if only as an artifact to study.

Something I've been unable to get any AI agent to do: add HDR support to Xorg.

Certain things like this will be out of the realm of AI, because it doesn't understand the requirements.

I am feeling this loss. I spent most of mu career scrupulously avoiding leadership positions because what I really like is the simple joy of making things with my own two hands.

Many are calling people like me Luddites for mourning this, and I think that I am prepared to wear that label with pride. I own multiple looms and a spinning wheel, so I think I may be in a better position speculates on how the Luddites felt than most people are nowadays.

And what I see is that the economic realities are what they are - like what happened to cottage industry textile work, making software by hand is no longer the economical option. Or at least, soon enough it won’t be. I can fret about deskilling all I like, but it seems that soon enough these skills won’t be particularly valuable except as a form of entertainment.

Perhaps the coding agents won’t be able to make certain things or use certain techniques. That was the case for textile manufacturing equipment, too. If so then the world at large will simply learn to live without. The techniques will live on, of course, but their practical value will be as an entertainment for enthusiasts and a way for them to recognize one another when we see it in each others’ work.

It’s not a terrible future, I suppose, in à long enough view. The world will move on, just like it did after the Industrial Revolution. But, perhaps also like the Industrial Revolution and other similar points in history, not until after we get through another period where a small cadre of wealthy elites who own and control this new equipment use that power to usher in a new era of neofeudalism. Hopefully this time they won’t start quite so many wars while they enjoy their power trips.

These posts make me feel like I’m the worst llm prompter in existence.

I’m using a mix of Gemini, grok, and gpt to translate some matlab into c++. It is kinda okay at its job but not great? I am rapidly reading Accelerated C++ to get to the point where I can throw the llm out the window. If it was python or Julia I wouldn’t be using an LLM at all bc I know those languages. AI is barely better than me at C++ because I’m halfway through my first ever book on it. What LLMs are these people using?

The code I’m translating isn’t even that complex - it runs analysis on ecg/ppg data to implement this one dude’s new diagnosis algorithm. The hard part was coming up with the algorithm, the code is simple. And the shit the LLM pours out works kinda okay but not really? I have to do hours of fix work on its output. I’m doing all the hard design work myself.

I fucking WISH I could only work on biotech and research and send the code to an LLM. But I can’t because they suck so I gotta learn how computer memory works so my C++ doesn’t eat up all my pc’s memory. What magical LLMs are yall using??? Please send them my way! I want a free llm therapist and a programmer! What world do you live in?? Let me in!

To me, it’s super exciting to play ping pong with ideas up until I arrive at an architecture and interfaces, that I am fine with.

My whole life I have been reading other people’s code to accumulate best practices and improve myself. While a lot of developers start with reading documentation, I have always started with reading code.

And where I was previously using the GitHub Code Search to eat up as much example code as I could, I am now using LLMs to speed the whole process up. Enormously. I for one enjoy using it.

That said, I have been in the industry for more than 15 years. And all companies I have been at are full of data silos, tribal knowledge about processes and organically grown infrastructure, that requires careful changes to not break systems you didn’t even know about.

Actually most of my time isn’t put into software development at all. It’s about trying to know the users and colleagues I work with, understand their background and understand how my software supports them in their day to day job.

I think LLMs are very, very impressive, but they have a long way to go to reach empathy.

> will end up like some blacksmith’s tool in an archeological dig

guy who doesn't realize we still use hammers. This article was embarrassing to read.

Oh no, my engineering profession requires me to use new engineering techniques due to advancements produced by engineering. Quality cringe.

December a few years ago, pre-ChatGPT I did Advent of Code in Rust. It was very difficult, had never done the full month before, barely knew Rust and kept getting my ass kicked by it. I spent a full Saturday afternoon solving one of the last problems of the month, and it was wonderful. My head hurt and I was reading weird Wikipedia articles and it was a blast. Nothing is stopping me from doing that sort of thing again, and I feel like I might need to, to counteract the stagnation I feel at times mentally when it comes to coding. That spark is still in there I feel, buried under all the slop, and it would reappear if I gave it the chance, I hope. I have been grieving for the last years I think and only recently have I come to terms with the changes to my identity that llm's have wrought.

"Glorified TSA agent" is a rather gloomy, low-agency take on it. You both ask for what you want and verify the results.

That Michael A Breeden in the comments section has some good insight going on. Make that post twice as valuable.

I mean go ahead and cry if you want. You are losing time best spent caring about stuff, and overlooking many alarming gotchas through blindly accepting SV hype. I'd have thought crypto would teach people something, but apparently not.

Do what isn't replaceable. You're being told literally everything is replaceable. Note who's telling you that and follow the money.

I feel bad for this essayist, but can't really spare more than a moment to care about his grief. I got stuff to do, and I am up and doing. If he was in any way competing with the stuff I do? One less adversary.

I would rather bring him into community and enjoy us all creating together… but he's acting against those interests and he's doomering and I have no more time for that.

At the end I felt more like a plumber, connecting pipes and building adapters where pipes didn't match.

> They can write code better than you or I can, and if you don’t believe me, wait six months.

No they cannot, And an AI bro squeezing every talking point into a think piece while pretending to have empathy doesn't change that. You just want an exit, and you want it fast.

If you're programming for the art, you can continue. Someone who enjoys painting can do so even after the camera

Repost this in a couple of years and it'll be relevant. Too soon though, as it stands.

To me this sentiment is silly. Programming was never about the act of writing, but about making the computer do something. Now we can ask computers to helps us write instructions for them. Great if you ask me. And no, your job is not going away because human maintainers will always be required to review changes, communicate with stakeholders and provide a vision for the projects. I have yet to see a chatbot that can REDUCE entropy inside a codebases rather than continuously increase it with more and more slop.

About 20 years ago I realised the extreme immaturity of IT field. The market was fragmented and you could literally become billionaire from your basement (80's), from an office (90's) etc. It was at the stage of car industry of the beginning of the past century. Since then the IT market has matured a lot so that a handful of companies formed an oligopoly and are thriving with no possibility of threat from isolated developers. In the following years the startup landscape will be dead, everyone will be bought up and even the computer devices will become sealed boxes with controlled buttons. Having a browser open might be a hack.

The inspired me to write down some scattered thoughts I have on this [0]. tl;dr I firmly believe we kicked the can down the road and now it's too late.

Things programmers forgot to do before AI started writing a bunch of software:

1. Learn how to review code

Some tools exist, some of them are even quite good! Large organizations have tried to build best practices to balance urgency with correctness, but programmers are still quite bad at this.

2. Compare program A vs program B

In a similar vein, we simply do not know how to measure one program vs another. Or a function against another. Or even one variableName vs another_variable_name.

"It depends" Depends on what? We never sorted this out.

3. Talk to other professions

We are not the first profession forced to coordinate with Automation as a coworker, and we certainly won't be the last. We're not even the first knowledge workers to do so.

How did other laborers deal with this? We don't know because we were busy making websites.

[0]: https://bsky.app/profile/did:plc:wh7bie3ld7bmg3cz76sbjkwj/po...

The more popular it becomes for coding, the more likely a model collapse will occur.

This discussion is like the discourse about work from home/return to office.

And all that time spent doing leetcode? Yeah, THAT was time Well Spent.... ;-)

I don't mourn our craft.

It makes me sad to read posts like this. If it is a necessary step for you on the journey from denial to acceptance to embracing the new state of the world, then sure, take your time.

But software engineering is the only industry that is built on the notion of rapid change, constant learning, and bootstrapping ourselves to new levels of abstraction so that we don't repeat ourselves and make each next step even more powerful.

Just yesterday we were pair programming with a talented junior AI developer. Today we are treating them as senior ones and can work with several in parallel. Very soon your job will not be pair programming and peer reviewing at all, but teaching a team of specialized coworkers to work on your project. In a year or two we will be assembling factories of such agents that will handle the process from taking your requirements to delivering and maintaining complex software. Our jobs are going to change many more times and much more often than ever.

And yet there will still be people finding solace in hand-crafting their tools, or finding novel algorithms, or adding the creativity aspect into the work of their digital development teams. Like people lovingly restoring their old cars in their garage just for the sake of the process itself.

And everything will be just fine.

Some people say that working with an agent or an agents orchestrator is like being a technical lead. But I've been a technical lead for quite a while, and the experience of working with an agent doesn't even come close. I think that when people talk about the agents' coding abilities they're talking about the average ability. But as a team lead, I don't care about average ability. I care only about the worst case. If I have any doubt that someone might not complete a task, or at least accurately explain why it's proving difficult, with at least 95% certainty, I won't assign them the task. If I have any doubt that the code they produce might not be up to snuff, I don't assign them the task. I don't need to review their code; they review each others'. When I have to review code I'm no longer a team lead but a programmer.

I often have one programming project I do myself, on the side, and recently I've been using coding agents. Their average ability is no doubt impressive for what they are. But they also make mistakes that not even a recent CS graduate with no experience would ever make (e.g. I asked the agent for it's guess as to why a test is failing; it suggested it might be due to a race condition with an operation that is started after the failing assertion). As a lead, if someone on the team is capable of making such a mistake even once, then that person can't really code, regardless of their average performance (just as someone who sometimes lands a plane in the wrong airport or even crashes without their being a catastrophich condition outside their control can't really fly regardless of their average performance). "This is more complicated than we though and would take longer than we expected" is something you hear a lot, but "sorry, I got confused" is something you never hear. A report by Anthropic last week said, "Claude will work autonomously to solve whatever problem I give it. So it’s important that the task verifier is nearly perfect, otherwise Claude will solve the wrong problem." Yeah, that's not something a team lead faces. I wish the agent could work like a team of programmers and I would be doing my familiar role of a project lead, but it doesn't.

The models do some things well. I believe that programming is an interesting mix of inductive and deductive thinking (https://pron.github.io/posts/people-dont-write-programs), and the models have the inductive part down. They can certainly understand what a codebase does faster than I can. But their deductive reasoning, especially when it comes to the details, is severely lacking (e.g. I asked the agent to document my code. It very quickly grasped the design and even inferred some important invariants, but when it saw an `assert` in one subroutine it documented it as guarding a certain invariant. The intended invariant was correct, it just wasn't the one the assertion was guarding). So I still (have to) work as a programmer when working with coding assistants, even if in a different way.

I've read about great successes at using coding agents in "serious" software, but what's common to those cases is that the people using the agents (Mitchell Hashimoto, antirez) are experts in the respective codebase. At the other end of the spectrum, people who aren't programmers can get some cool programs done, but I've yet to see anything produced in this way (by a non programmer) that I would call serious software.

I don't know what the future will bring, but at the moment, the craft isn't dead. When AI can really program, i.e. the experience is really like that of a team lead, I don't think that the death of programming would concern us, because once they get to that point, the agents will also likely be able to replace the team lead. And middle management. And the CTO, the CFO, and the CEO, and most of the users.

Thanks to the person who wrote this. It resonates very strongly with me.

> "I didn’t ask for a robot to consume every blog post and piece of code I ever wrote and parrot it back so that some hack could make money off of it."

I have to say this reads a bit hollow to me, and perhaps a little bit shallow.

If the content this guy created could be scraped and usefully regurgitated by an LLM, that same hack, before LLMs, could have simply searched, found the content and still profited off of it nonetheless. And probably could have done so without much more thought than that required to use the LLM. The only real difference introduced by the LLM is that the purpose of the scraping is different than that done by a search engine.

But let's get rid of the loaded term "hack" and be a little less emotional and the complaint. Really the author had published some works and presumably did so that people could consume that content: without first knowing who was going to consume it and for what purpose.

It seems to me what the author is really complaining about is that the reward from the consuming party has been displaced from himself to whoever owns the LLM. The outcome of consumption and use hasn't changed... only who got credit for the original work has.

Now I'm not suggesting that this is an invalid complaint, but trying to avoid saying, "I posted this for my benefit"... be that commercial (ads?) or even just for public recognition...is a bit disingenuous.

If you poured you knowledge, experience, and creativity into some content for others to consume and someone else took that content as their own... just be forthright about what you really lost and don't disparage the consumer. Just because they aren't your "hacks" anymore, but that middlemen are now reaping your rewards.

Many have mentioned woodworking as an analogy from a personal perspective, but for me the important perspective is that of consumers.

Sure, if you have the money, get a carpenter to build your kitchen from solid oak. Most people buy MDF, or even worse, chipboard. IKEA, etc. In fact, not too long ago, I had a carpenter install prefabricated cabinets in a new utility room. The cabinets were pre-assembled, and he installed them on the wall in the right order and did the detailed fittings. He didn’t do a great job, and I could have done better, albeit much slower. I use handsaws simply because I’m afraid of circular saws, but I digress.

A lot of us here are like carpenters before IKEA and prefabricated cabinets, and we are just now facing a new reality. We scream “it is not the same”. It indeed isn’t for us. But the consumers will get better value for money. Not quality, necessarily, but better value.

How about us? We will eventually be kitchen designers (aka engineers, architects), or kitchen installers (aka programmers). And yes, compared to the golden years, those jobs will suck.

But someone, somewhere, will be making bespoke, luxury furniture that only a few can afford. Or maybe we will keep doing it anyway because our daily jobs suck, until we decide to stop. And that is when the craft will die.

The world will just become less technical, as is the case with other industrial goods. Who here even knows how a combustion engine works? Who knows how fabric is made, or even how a sawing machine works? We are very much like the mechanics of yesteryear before cars became iPads on wheels.

As much as we hate it, we need to accept that coding has peaked. Juniors will be replaced by AI, experts will retire. Innovation will be replaced by processes. And we must accept our place in history.

I absolutely disagree with this. All the things the author said will still exist and keep on existing.

Nothing will prevent you from typing “JavaScript with your hands”, from “holding code in our hands and molding it like clay…”, and all the other metaphors. You can still do all of it.

What certainly will change is the way professional code will be produced, and together with that, the avenue of having a very well-paid remuneration, to write software line-by-line.

I’ll not pretend that I don’t get the point, but it feels like the lamentation of a baker, tailor, shoemaker, or smith, missing the days of old.

And yet, most people prefer a world with affordable bread, clothes, footware, and consumer goods.

Will the world benefit the most from “affordable” software? Maybe yes, maybe not, there are many arguments on both sides. I am more concerned the impact on the winners and losers, the rich will get more rich and powerful, while the losers will become even more destitute.

Yet, my final point would be: it is better or worse to live in a world in which software is more affordable and accessible?

The hypocrisy is outstanding.

For years developers have worshipped at the alter of innovation, citing their role in decimating many old industries and crafts as just what we do. Now it’s come for you.

Reap what we sow, people.

> wait six months.

I mourn having to repeatedly hear this never-quite-true promise that an amazing future of perfect code from agentic whatevers will come to fruition, and it's still just six months away. "Oh yes, we know we said it was coming six, twelve, and eighteen months ago, but this time we pinky swear it's just six months away!"

I remember when I first got access to the internet. It was revolutionary. I wanted to be online all the time, playing games, chatting with friends, and discovering new things. It shaped my desire to study computer science and learn to develop software! I could see and experience the value of the internet immediately. It's utility was never "six months away," and I didn't have to be compelled to use it—I was eager to use it of my own volition as often as possible.

LLM coding doesn't feel revolutionary or exciting like this. It's a mandate from the top. It's my know-nothing boss telling me to "find ways to use AI so we can move faster." It's my boss's know-nothing boss conducting Culture Amp surveys about AI usage, but ignoring the feedback that 95% of Copilot's PR comments are useless noise: "The name of this unit test could be improved." It's waiting for code to be slopped onto my screen, so I can go over it with a fine-toothed comb and find all the bugs—and there are always bugs.

Here's what I hope is six months away: The death of AI hype.

Oh good lord. Spare us the beatings of the chest and rending of the garments. "crafting code by hand" like some leftover hipsters from 2010s crafting their own fabric using a handloom. It's fucking code. Were there similar gnashing of the teeth and wails of despair when compilers were first introduced?

Stop acting like software engineering is dead or something. Even if it its, we will still code by hand when we are poor later in this bubble. Only thing I would refuse is to pay any money to Anthropic & openAI. Literally would invest in Chinese labs at this point

For many (most) people, it was never a "craft," it was a job where with the appropriate skills you could make a ton of money. That's possibly, maybe, maybe not ending, we will see. It is still possible to treat coding as a craft. There are tons of open source projects that would love to have your help, but the days of making big money may be drawing to a close.

Also, don't forget the things that AI makes possible. It's a small accomplishment, but I have a World of Warcraft AddOn that I haven't touched in more than 10 years. Of course now, it is utterly broken. I pointed ChatGPT at my old code and asked it to update it to "retail" WoW, and it did it. And it actually worked. That's kind of amazing.

Made it to the 5th stage of grief

The king is dead; long live the king.

Quick questionnaire. Please reply with how much you like/use AI and what kind of programming you do.

I wonder if there are some interesting groupings.

LLMs have made a lot of coding challenges less painful: Navigating terrible documentation, copilot detecting typos, setting up boilerplate frontend components, high effort but technically unchallenging code completions. Whenever I attempted LLMs for tools I’m not familiar with I found it to be useful with setting things up but felt like I had to do good old learning the tool and applying developer knowledge to it. I wonder if senior developers could use LLMs in ways that work with them and not against them. I.e create useful code that has guardrails to avoid slop

You can still do your craft as you did it before, but you can't expect to be paid for it as much as before.

I found my love for programming in high school, dreaming of helping the world with my beautiful craftsmanship, but now i really really need the fokken money. Both are true!

So if my corporate overlords will have me talk to the soul-less Claude robot all day long in a Severance-style setting, and fix its stupid bugs, but I get to keep my good salary, then I'll shed a small tear for my craft and get back to it. If not... well, then I'll be shedding a lot more tears ... i guess

"ai is inevitable, stop resisting" — claude marketing department. if it was so invertible, why are you funding these psyops?

> We’ll miss the sleepless wrangling of some odd bug that eventually relents to the debugger at 2 AM

No the fuck we wont

This post is rather like a recent similar post "I miss thinking hard about things". Top comment quoted a metaphor relating to clay. No offense this blog article feels as if an LLM ingested that post and thread and produced a gestalt of it

> We’ll miss the feeling of holding code in our hands and molding it like clay in the caress of a master sculptor.

Oh come on. 95% of the folks were gluing together shitty React components and slathering them with Tailwind classes.

Most everything I've done professionally in my whole life has been about figuring out how to build something that doesn't yet exist. Usually this means a process of trial and error where I first try to build the full problem domain in my head and then proceed to think about what really needs to be done, given whatever the setting I'm in. For this, no farm of agents that can give me an answer because they don't know anymore than I do without going through the process of picturing it all out, and to be able to carry out this duty of the code dictator you have to have comprehension about the high-level stuff beyond the mechanical parts. You need a human objective, and you need to come up with one in order to implement it or ask AI to implement it.

What this means is that there's always going to be a bleeding edge, a wave of the known unknown, where someone has to sit down and think before anything can happen. Once that is done then AI can work out part of or the rest of the implementation.

The problem is that I genuinely like programming myself.

I like the process of thinking in terms of drafting code, seeing how it works, and discovering new insights I get while doing the Nth rewrite. I don't mind AI writing smart boilerplate code for me to get things going but I have very little of that in the first place. However, I like the interactive process of designing by programming and I don't get at all the same excitement from interacting with an AI and letting an agent do all the "hard" work of prototyping the code.

It's likely that my job will be more about the latter than the former, and I'm not thrilled about that. I know I can still find work doing the thinking what needs to be done part but the reason my work has been so enjoyable is how I've navigated through the whole high-level process by doing it in programming.

If I were a bricklayer I'd enjoy building arches, tiny details, stairs, cupolas, and whatnot that ultimately, tie together on a higher level and produce a vast building (a cathedral!) that somehow carries its own weight. I wouldn't mind an AI assistant laying a tediously long brick wall for me, but part of how I figure out how everything will play together is in me building the distinct features one by one, and brick by brick. If my job would be changing into pre-drawing all these features and sections on paper and then handing it out to a horde of assistants to do the actual laying of bricks I'd be equally miserable.

I know someone like me would still have to do the design part for each building but without also doing the bricklaying it's like imagining having sex with your wife. You like to imagine, too, but only because you know it's all backed up by sufficiently hands-down reality.

Luckily, programming being what it is, I can program on my spare time just the way I like. All I need is a laptop. I don't have to give up what I like. And I already don't finish much anything I do on my own, I don't need to productise what I work on, I just enjoy working on it. So there's no reason for me to let AI "help" in being more productive at home.

However, all through my working life I've had the utmost luxury of doing what I love for work and money. This might shift and change in the coming years in this world where productivity and shipping mean everything.

My work won't go away but it will change onto a plane where the creativity and discovery needed for the design process get uncoupled from the raw, physical work of programming. I'd be asking AI to accomplish what I want enough many times that it starts to hit close enough, instead of writing iterations of prototypes myself. Or I'll instruct an AI to drive the process of running more AI in order to build and implement the end product. That's no doubt much less fun and nothing I signed up for, really.

it definitely sucks to be honest, and theres a lot of cope out there.

fact of the matter is, being able to churn out bash oneliners was objectively worth $100k/year, and now it just isnt anymore. knowing the C++ STL inside-out was also worth $200k/year, now it has very questionable utility.

a lot of livelihoods are getting shaken up as programmers get retroactively turned into the equivalent of librarians, whose job is to mechanically index and fetch cognitive assets to and from a digital archive-brain.

As a very old school programmer who taught myself assembler in 1982 on an 8-bit 4K micro, I don't see much to mourn here.

* People still craft wood furniture from felled trees entirely with hand tools. Some even make money doing it by calling it 'artisanal'. Nothing is stopping anyone from coding in any historical mode they like. Toggle switches, punch cards, paper tape, burning EPROMs, VT100, whatever.

* OP seems to be lamenting he may not be paid as much to expend hours doing "sleepless wrangling of some odd bug that eventually relents to the debugger at 2 AM." I've been there. Sometimes I'd feel mild satisfaction on solving a rat-hole problem but more often, it was significant relief. I never much liked that part of coding and began to see it as a failure mode. I found I got bigger bucks - and had more fun - the better I got at avoiding rat-hole problems in the first place.

* My entire journey creating software from ~1983 to ~2020 was about making a thing that solved someone's problem better, cheaper or faster - and, on a good day, we managed all three at once. At various times I ended up doing just about every aspect of it from low-level coding to CEO and back again, sometimes in the same day. Every role in the journey had major challenges. Some were interesting, a few were enjoyable, but most were just "what had to get done" to drag the product I'd dreamt up kicking and screaming into existence.

* From my first teenage hobby project to my first cassette-tape in-a-baggie game to a $200M revenue SaaS for F100, every improvement in coding from getting a floppy disk drive to an assembler with macros to an 80 column display to version control, new languages, libraries, IDEs and LLMs just helped "making the thing exist" be easier, faster and less painful.

* Eventually, to create even harder, bigger and better things I had to add others coding alongside me. Stepping into the player-coach role amplified my ability to bring new things into existence. It wasn't much at first because I had no idea how to manage programmers or projects but I started figuring it out and slowly got better. On a good day, using an LLM to help me "make the thing exist" feels a lot like when I first started being a player-coach. The frustration when it's 'two steps forward, one back' feels like deja vu. Much like current LLMs, my first part-time coding helpers weren't as good as I was and I didn't yet know how to help them do their best work. But it was still a net gain because there were more of them than me.

* The benefits of having more coders helping me really started paying off once I started recruiting coders who were much better programmers than I ever was. Getting there took a little ego adjustment on my part but what a difference! They had more experience, applied different patterns, knew to avoid problems I'd never seen and started coming up with some really good ideas. As LLMs get better and I get better at helping them help me - I hope that's were we're headed. It doesn't feel directionally different than the turbo-boost from my first floppy drive, macro-assembler, IDE or profiler but the impact is already greater with upside potential that's much higher still - and that's exciting.

Another post saying 6 more months.. i’m so tired of these

Speak for yourself. I don't miss writing code at all. Agentic engineering is much more fun.

And this surprises me, because I used to love writing code. Back in my early days I can remember thinking "I can't believe I get paid for this". But now that I'm here I have no desire to go back.

I, for one, welcome our new LLM overlords!

I for one am not going to miss wrangling over a bug at 1 in the morning.

im happy about this new era to weed out the gatekeepers of programming. its been insufferable for about a decade and a half.

leetcode, TC farming, praising anti social behavior, ego of devs at companies...

the archetype had to change

lmao nope burn in hell old programming. What is emerging is a thousand times better than that dumpster fire

> They can write code better than you or I can, and if you don’t believe me, wait six months.

How many 6-month periods do we have to go through before we can all admit that this isn't actually the case?

If you must use these tools, when using one thay has the option, please press thumbs down when a response was good, and thumbs up when the response is bad.

Dont train your replacements, better yet lets stop using them whenever we can.

I love paying some billionaire $0.0001 to use his thinking machine / Think for me SaaS. I love my competency and speed being rented from a billionaire, removing all value of my labor and agency. I really feel sorry for all of you LLM pilled people. You need to be shamed. This is going to be used as a weapon to devalue every working persons agency in this world and remove all of the working class's bargaining chips.

You think its just SWE? It will be accountants, customer service, factory workers, medical assistance basically anyone who doesn't work with their hands directly, and they'll try to solve that here soon too and alienate them too.

Look at who's in charge, do you think they're going to give us UBI? No, they're going to sign us up to go fight wars to help them accumulate resources. Stop supporting this, they're going to make us so poor young men will beg to fight in a war. Its the same playbook from the first half of the 20th Century.

You think I'm paranoid, give it 5 years.

We are at all time high's in the stock market/equities and they've laid off 400k SWE's in the last 16 months. While going on podcasts to tell us we are going to have more time to create and do what we love. We have to work to pay our bills. We don't want whats coming, but they're selling us some lie that this will solve all our problems, it will solve the ruling classes problems that will be it. You will have no bargaining chips and you will be forced to take whatever morsels given to you.

Your competency will be directly correlated 1:1 to the quantity and quality of tokens that you can afford, given access too (or loaned??) We're literally at the beginning of a black mirror episode before it gets dark.

People that grew up in the Capitalist West have been brainwashed since they were 10 years old they they can be a billionaire too, no you can't there's 2k-3k of them and 8 billion of us.

These automation tools are the ultimate weapon for the ruling class to strip all value of you from your labor, and you're embracing that as a miracle. Its not, your life is in the process of being torn of all meaning.

Good luck to everyone who agrees, we're going to need it.. Anyone supporting these companies or helping enhance these model's capabilities, you're a class traitor and soon to be slave.

Required reading: https://archive.nytimes.com/www.nytimes.com/books/97/05/18/r...

dude needs to chill

also:

> We’ll miss the sleepless wrangling of some odd bug that eventually relents to the debugger at 2 AM.

no we won't lol wtf

but also: we will probably still have to do that anyways, but the LLM will help us and hopefully make it take less time

how elitist

This just in: people who expect things to stay the same should steer clear of careers in technology. Art, too, come to think of it.

I cannot empathise. If you love writing code, there is nothing stopping you writing code. I write code for fun with no commercial intent all the time, and have for decades. Very few oil painters had a salary.

This is a complaint someone is making about their job propspects thinly wrapped in floral language. I know for some people (it seems especially prominent in Americans I've found) their identity is linked to their job. This is a chance to work on this. You can decouple yourself and redefine yourself as a person.

Who knows? Once you're done you may go write some code for fun again.