Hello programmers! I’ve been making a living for a couple of decades figuring out stuff for people and coding whatever will make them happy. It’s fun. I’ve grown to prefer the people to the coding but I still get a kick out of the whole thing.
The problem is: at the job we’re being asked to pivot to vibe coding. At this point the order is to use it for all code, read the slop carefully, argue with the machine for way too long and only then open that PR. Massive productivity gains are expected. I’m appalled from an ethical, philosophical and professional standpoint. We’re a non-profit ngo ffs. My colleagues are fine with it. If I get even slightly critical the group goes awkwardly silent and I get comments on how all the programmers they know from other places don’t code anymore.
If this is how it goes from now on, I’m out. Is it really though? Are all businesses switching to vibe coding? Have some of you started switching career? I think I want to go to meatspace. Hand-made coding could make a cute hobby.
I am working in a very small company as a software dev and write everything by hand. I wrote some weeks ago in a different thread here that I tried all of the LLM powered coding tools and found all of them subpar.
If this company goes downhill or I lose the job otherwise, I will work something entirely different.
My very arrogant point of view is that these “vibecoding” tools are being celebrated by people who were making shit software before and now make software that is slightly less shit in what they perceive is the same amount of time, but I suspect it takes them actually longer. Meanwhile, they don’t grow their programming skills at all.
people who can suddenly “”“make”“” things like minecraft mods are also a part of those celebrating, though i assume they’re a lot less vocal about it
and god i wish that was all it was used for
From what I can observe, yeah, pretty much the entire industry has shifted to LLM code generation. I’m sure (or at least I hope) there are still some shops that go out of their way to keep working like we did, you know, months ago. But beyond that… I’m not even sure the general public understands how quickly and how fully LLM codegen has become the norm. It’s grim.
I’m a systems architect who uses agentic coding to get around some executive dysfunction challenges with writing my own code. I learned about the dysfunction late in life; maybe meds will eventually get me to a point where I can consistently work on a personal code project without a looming deadline or social contract in the loop. For now I am part of the problem. Hi!
I look at agentic coding as a firearms problem:
- it lowers the barrier to execution: anyone with sloppy aim can point and pull the trigger, and do a lot of harm in the process
- both perform better in the hands of people with discipline and experience, but that doesn’t change the fact that LLMs and guns have massively proliferated and any idiot can get their hands on one (or that shitty people benefit from mass producing them)
- the industrial complex producing guns profits from murder, and the industrial complex producing datacenters profit from murder of the environment and attempts to eliminate jobs
- the same countries who treat gun control as an anathema tend to view AI control as an anathema
- analogies suck, they’re never perfect but they get the conversation started
That said, most of the people who hate AI also don’t know a damn thing about how they’re used correctly. It’s also not their fault. Sure, they’re under no obligation to be an expert about the thing that’s harming them. But there’s a bigger reason: the companies pushing this shit benefit from a world stage where people use them wastefully and inefficiently.
The short list of things most people don’t get:
Spoiler'd because most people also won't give a fuck.
- A LLM is like a ten year old with a doctorate who wants to impress you. It can answer random trivia off the top of its head, but it’s going to make shit up when it doesn’t know the answer.
- Never trust a LLM answer that was produced without a web search or web fetch. Pretend you’re that obnoxious algebra teacher who rolled their eyes at your calculator and told you to show the work during exams. Work for a LLM is tool calls: web searches and webpage fetches. Tell it to research documentation and best practices before you start. If you don’t, the answer you get will be dated by 1-2 years in the best case, and a confidently wrong hallucination at worst. (unfortunately you’re still at risk of being fucked because frontier labs use webpage summarization in these tool calls, and that summarization adds another vector for hallucinated shit; lovely)
- The baseline skills of the operator are more important than anyone gives credit. Your results scale to your own knowledge level and your use of the LLM to extend into adjacent subject matter. Employers don’t want to hear this: they want employees to go away. Frontier LLM labs don’t want employers to hear this: they need employers to buy into the delusion of eliminating their work force, and also financially benefit from people using the tools wastefully. (token burn = profit = the heat death of the universe as an afterthought, but that’s the next generation’s problem) Casual LLM critics don’t want to hear this: confidently incorrect echo chambers are the comfort food of anyone with an axe to grind about what they hate.
- Consistent results require controlling the probability space so that your pipeline swats the hand of the LLM when it’s lazy or generating slop. For coding, that means using languages or tooling that prevent major classes of problems to begin with (Rust, Go, Typescript) or toolchains that bitch about sloppy code (pytest-ruff + pytest-ty for Python). It means extensive unit tests that cover expressions of functionality, and knowledge of how to prompt out bad behaviors. (don’t rewrite tests to conform to “preexisting bugs”, always align tests to higher standards, etc.)
I could add another ten bullet points here, but I’m not here to prove that LLMs are capable of being useful. I’m here to demonstrate that it requires knowledge of a problem domain to get the best results. That means you can’t get rid of employees familiar with the knowledge domain. You need senior employees to get the best results, and you need junior employees so that they can eventually become senior employees. Same as it ever was. But because we’ve got useless fucking politicians who are too busy lining their own pockets instead of, I don’t know, regulating a worldwide memory shortage caused by a handful of companies, the whole thing is cannibalization of human society all the way down.
We could have scaled this to operate in the consumer space at a fraction of the energy costs (Apple SOC style versus energy inefficient hyperscaling) and used it as something to generate value by augmenting the capabilities of individual workers instead of trying to replace them wholesale. It only makes sense that it should have benefitted everyone considering this tech was built on shoving libraries worth of human achievement into a training pipeline (often without permission), right?
Nah. They profit more if we don’t own our own compute and everyone is encouraged to burn energy on treating the LLMs like all-knowing oracles instead of research tooling. They control the access to creations built on stolen knowledge. But hey, if everything else gets burned down in the process at least they come out on the other side rich and bailed out.
If you work for a company that prioritizes profit over quality - yes.
I no longer work in government IT, but from what I have heard from old colleagues, quality is still prioritized there and they are not forced to vibe code.
I read a retrospective from another team yesterday. They were preparing a demo for leadership and apparently it didn’t go so well. There were many reasons like last minute changes in requirements but one line was interesting for this discussion: The use of the LLM agents caused confusion about who is fixing which bug and taking on which tasks. The agents act like they’re alone in the codebase and keep adding and changing code broadly whenever they see an opportunity for improvement. And that often without specifically being prompted to do so. Now if you have multiple team members collaborating on a project with each their agents, chaos ensues because the agents are not coordinating and the diffs/commits become insane and the team doesn’t fully understand them.
This actually results in a necessity to test more and longer because nobody knows which bugs have really been fixed now and rapidly preparing for a release or a demo becomes more difficult than before.
In my opinion an LLM can be very useful to rapidly prototype something. Especially with an API or language you’re not familiar with. But for Production it’s still a huge risk to push code you don’t understand.
Almost everyone in the industry worth their weight is using LLMs now. Top tier models create really solid code nowadays.
Almost everyone in the industry worth their weight is using LLMs now. Top tier models create really solid code nowadays.
This is not my experience. What worries me, is that I see posts like this all across the internet:
- All the good programmers are using LLMs. If you are not using agents for coding, you’re a bad programmer.
- Frontier LLMs are wonderful. You are just using the wrong model.
Anyway, if it works well for you, then I am happy for you. If you feel the need to evangelise, I think it would be more effective if you include some of your personal experience and include what is bad along with what is good.
Here is my personal experience from this Thursday:
- Gemini recommended a Kubernetes service to me (nice choice).
- The Helm config it gave had big problems. Settings were hallucinated and it used a feature that did not exist.
- After a day of doubling down on errors, I read a Github issue that said the Helm chart did not support the feature I needed.
- On Friday, I wrote my own Helm chart for the program from scratch. Thankfully, this is a skill that I have not lost. Its small, simple and I understand it.
- Total score this week: 1 day lost to AI.
The issue there is that yo are using a model that is well behind the frontier for coding (google havent released a new pro model for a year now) and using the chatbot interface. When people talk about using LLMs for coding it is using agents that can take action, see when things go wrong, do their own research and correct their failures. If you had been doing that I doubt you would have lost a day having to fix things up.
You still might have had to do the work by hand, thats an outcome that can happen, but it would have got to the point of “this service cant do that” by itself without time wasting back and forth with you.
We’re going towards a world where we can’t trust code anymore than we can trust images or videos. People all over the place are entrusting their entire jobs to AI and consumers have to deal with the output.
Software engineering has always been bad. People just copy pasted stack overflow. Now an LLM copies stack overflow and the documentation.
If you seriously think that’s what software engineering was and still is, I feel sorry for you. You think opensource software is just a series of copy-pastes from stack overflow with no understanding?
I write code. There is skill to being able to read documentation. I love tools like predictive IDE’s. Search engines augmented my ability to code with sites like stackoverflow. LLM’s are a continuation of that trend. Often the output is broad strokes that I need to refine. At my workplace, humans are required even with the tool improvements.
One question I have for the slop advocates in this thread is this - do you not like programming? Did you fall for the „learn to code“ meme and just learned the bare minimum to get a paycheque working on web dashboards?
Like why would you voluntarily cut out the most enjoyable part of the job, and go around scoffing at people who don’t want to do that?
Productivity is my boss’s problem, not mine, and I’d rather go and try and farm geese (I hate geese) than spend my whole day reading LLM generated code.
There’s nothing to prevent you from writing code in an IDE while the torment nexus TUI churns out whatever monster misfeature request was waiting for you in your Jira queue this morning.
In fact I’d recommend doing the code writing part. It helps, even if it increasingly becomes a rounding error in your overall output.
It’s bleak as shit right now. God why couldn’t this have waited until after my retirement.
The responses to your question makes me fear for the future of products.
I use LLMs for scaffolding, boilerplate stuff, and basically things I can trust a junior with. For anything else, I have been forced to write it myself or rewrite it because the agents were so goddamn awful at it.
Reading that people even think programming has nothing to do with code and is just solving problems is just nuts to me. Code is a spec for a problem solution. If someone thinks the spec is neither important nor that being able to understand the spec is important… It’s like saying “I don’t care how this bridge is designed, it’s built”. Say that again and tell me it doesn’t sound nuts.
It explains a lot about the industry before the advent of LLMs. I always wondered why Node.js took over in web dev so quickly and that so many people only care about having something done makes sense. I just wish they weren’t destroying one of my few favourite things about humanity (Open Source Software) while they wallow in their own misery.
LLM won’t destroy OSS for me. It will actually be the only thing left to use because proprietary software will be the first place for LLMs to take over.
Writing code was never “the job”, that is just the trivial typing part you do when the thinking and designing ends. The job is solving problems (which I enjoy very much thanks.)
This is like people who refuse to program in another language because they are “a .NET programmer”. That just tells me you’re a coder not an engineer. In my career I’ve been paid good money to write C51 assembler, C, C++, Tcl, Limbo, Java, C# and Rust (roughly in order) - LLMs are just what’s next.
I do enjoy programming, but I prefer “getting stuff done”. While I enjoy programming, I’d rather just get it done and move on. I like the results of having programmed.
Productivity is my problem, because I build stuff for myself, not just for my boss.
I’m far from a slop advocate, so this is more of a devil’s advocate point: plenty of code writing is not enjoyable.
If I’m working on my own project, to my own standards, writing exactly what I want to write I’m usually going to get enjoyment out of it. That’s pretty commonly not the case when writing code for an employer. Maybe the tech stack sucks, the product is inane or worse, you think the feature is a dumb idea but have to do it anyway, you’re bending over backwards to work around tech debt that you’re not allowed by management to fix, or you have to appease incompetent/out of touch architects or tech leads who presume to tell you how to do your job. I personally don’t enjoy writing that code very much. At a certain point it’s almost like nails on a chalkboard if you genuinely enjoy programming for its own sake: you know what good would look like, you know how far away what you’re working on is from good, and you feel sad at all the organizational inertia you’d need to overcome to get to good or, choosing not to do that, at compromising your standards. That bugs me, at least, and at a certain point makes it hard to even start certain work projects.
LLMs can make this at least bearable. Rather than spending hours looking into the change yourself, writing all of the code, fighting the shitty test framework and swearing at the past engineers who made it so bad, you let the robot figure it out and review its work. The result may still suck, but it was going to suck if you wrote it by hand too, for reasons largely out of your control. You can’t be fully hands off, and you still have to deal with the things you don’t like to get a good result, but you put yourself a step away from what bothers you and by doing so make it a little more pleasant. And, when you find work that’s actually fun, interesting, or rewarding, you just cherry pick that for yourself. I’ve grown to appreciate them for this reason. I have a lot less dread for the nails on a chalkboard work than I used to, anyway.
When my AI fucks up code I sometimes think “I probably would have fucked that up too”. And then I make it write 50 more tests in 2 minutes so it doesn’t happen again. The ROI in time saving is too tantalizing for me to be a meat-only code monkey.
It does really need all those tests. But they probably should have existed anyway and I’m damn sure most devs out there weren’t going to be so obsessive with coverage.
I like to code. I like tools. There is a trap in using the LLM tool for everything. A Maslow’s hammer and everything looks like a nail type of trap.
Programming was never enjoyable to me.
Making things is. LLMs allow you to be extremely pragmatic and enjoy making things or troubleshooting things without tedious line by line nature.
Work at a saas company in silicon valley, pretty much all the code being written is from an LLM here. Everything is still reviewed by humans, and a bunch of rounds of LLM reviews as well though.
Productivity has gone up as far as I can tell, we are shipping more features with a smaller staff. Support cases also seem to be going down, at least ones that reach an engineer like me since a lot are being resolved earlier in the escalation chain by a support technician, or just the customer talking to the built in helper LLM in the product.
Our token spend is pretty high though, looking like I’m at $1,200 on the month so far, so there’s that but don’t hear to many complaints from the finance department yet…
I’m liking the transition, as I have less work now. My day is a lot more waiting on the AI to complete a task or waiting for a human reviewer to do a pass on the code so I can spend more of my time doing important things, like scrolling on lemmy.
As for the future, I don’t think this is going away as much as everyone else in this thread seems to be wish casting for that to happen. The coding agents and models are only getting better, and tokens will only get cheaper as the data centers come on line. Long term I’ve been looking into becoming an electrician, but for now the economics don’t make sense with my current salary, so sticking with it until they lay me off…
I’m liking the transition, as I have less work now. My day is a lot more waiting on the AI to complete a task or waiting for a human reviewer to do a pass on the code so I can spend more of my time doing important things, like scrolling on lemmy.
Maybe it’s just me but this sounds like torture to me. I have enough progress bars to wait for in my work day and I would rather do something else with my life if waiting and reviewing generated code becomes the bulk of it. It’s good to see that some people do seem to enjoy doing it so maybe there’s a chance I don’t have to.
I’m with you. Nearly same story, finance sector, east coast. It’s fun. It’s relaxing and I tinker with so many projects I never would have had the time for.
This comment condensed all the thoughts and emotions of an entire episode of Black Mirror into a few paragraphs
Just have it do the tedious parts while you do the high-level planning. The interview with Matt Pocock on Pragmatic Engineer (podcast) discussed this approach in the latest episode. If I had to guess, I’d say that part of the conversation happened in the last 30 or 40 minutes of the interview, but the whole thing was interesting.
My software dev friend was told to use Claude to speed up his code writing. He finds it great but it requires review.
The problem with cast coding a bunch of vibe coded apps is you have to maintain the vibe coded apps.
As a developer of over two decades, ive seen ai aka llms good at:
- Creating small MVPs to prove something is possible as software
- Stealing and using other peoples code to make said apps.
- Quick very small directed scripts.
- Making up funny random shit. Such as aidungeon.
- Using up ram
And its mediocre at a lot of things.
And straight up bad at other things such as cooking and instructions that need to be followed to the letter.
It’s not going away.
Everyone who complains about the quality of the code it produces is overlooking one important fact: 90% of the code developed by the software engineering profession is utter dogshit, pasted together by people whose understanding just about extends to typing in stackoverflow.com by banging their head against they keyboard.
LLMs are not replacing artisinal hand-crafted code developed by a skilled engineer from a top-10 university, it’s replacing the utter shite produced by the people with an NVQ from Timbuctoo Community College that actually powers the vast majority of websites and enterprise software.
And it’s fucking good at it.
Look, when I started my career I used to hand-write assembly code. I used to personally review the output of the Keil C51 compiler because its optimiser was objectively dogshit. Nobody is doing that any more - partly because the compilers are better than programmers at writing assembler these days, and partly because we have so much CPU and memory we don’t know what to do with, so who cares if they aren’t. Both apply to LLMs, and we will soom be at a stage where reviewing - or even caring - about the quality of LLM produced code is as anachronistic as thinking you need the check the compiler’s output. (TBF, Clang/LLVM for AVR still produces dogshit assembler, but that’s by the by.)
Just as assembler did before, high-level code is going to become just another intermediate representation that isn’t really important; what will matter are the prompts and context you give to the LLMs. And there’s still a hell of a lot of engineering skills that need to go into that.
Software Engineers - the real ones - have nothing to fear from LLMs, and in many ways it’s much more satisfying working with an LLM than a team of barely skilled “coders”. The barely skilled coders who wouldn’t know malloc from a semaphore though? They’re fucked.
God I have þoughts about your post, but I’m not typing all of it out on my stupid phone, so maybe if I remember when I’m at a computer þis comment will serve as a placeholder.
Just as assembler did before, high-level code is going to become just another intermediate representation that isn’t really important; what will matter are the prompts and context you give to the LLMs.
Ah the old “LLMs are just a new layer of abstraction” argument. No they aren’t, and won’t be any time soon.
You mean that “write this service make no mistakes” executed with a LLM-as-a-service where the underlying model updates every week and the model itself is deprecated in 8 months isn’t a good idea for a source code representation?
Nah, it definitely is. It’s called job security.
On the first half, that’s not how it’s done; the input is a much more detailed description of design and architecture coupled with TDD.
On the second half - I happen to agree, that’s why I think the OpenAI/Anthropic business models are doomed to fail. But that’s where locally deployed models become key - predictable output.
They don’t need to be better than the best programmer in the world to change how 90% of code is written, they just need to be better than a junior dev from Timbuctoo. And feasably-deployable open models crossed that bar around Q1 this year and are only getting better (Qwen3.8-Flash-Next is another huge leap forward.) We may not be quite there yet, but the destination is now inevitable; adapt or find a new career.
On the first half, that’s not how it’s done; the input is a much more detailed description of design and architecture coupled with TDD.
Yes, I have these coworkers too. I’ve seen what they look like, and it doesn’t change my point.
But that’s where locally deployed models become key - predictable output.
Local controlled models help, yes. That isn’t what business do, though. And even for local models, the models are still stochastic as fuck. Maybe that changes in the future, in which case I’d reconsider that view (though natural language is still a terrible interface for telling a computer precisely what to do).
We may not be quite there yet, but the destination is now inevitable; adapt or find a new career.
How many years has it been now? Mythos was the end of the world, Fable had to be locked down by the US government, Astra is going to cause humanity to go extinct, and yet here we still are today with vibecoded, mass-produced slop.
I’m still designing and writing all code by hand. I don’t know that any of my coworkers are and know that many definitely are not. Our company does have copilot reviews built into CI and, at least for now, requires human review as well.
There’s a long history of codegen fads. This isn’t the first one. (Though it does have many unprecedented aspects.)
Here are some thoughts to bring you comfort:
- This can’t last. Everything from the economics of the vendors, to the inevitability of customers blowing their limbs off, to the diminishing returns on an AI-heavy process, to the depletion of the talent pool that makes responsible supervision/auditing possible, to impending model collapse… It has to end eventually. It may take a while, but it will happen.
- In the meantime, with tokens being sold way below cost, you can get a helluva deal on some cheap custom software if you’re smart about where and how you use it. I’m “encouraged” to use AI at work, and 90% of my use is for peripheral tools to make my own workflow easier — not load-bearing production code.
- It makes a fantastic cover for skunkworks projects. If you have a thing you’ve been wanting to do but can’t get it through the front door, use it as a “test project” for AI. You don’t have to actually vibe code it. Have Claude write some docs or commit messages or some esoteric test suite. Just something you can point to.
What is the last fad that the entire industry adopted in 9 months?
If cost per intelligence is going down, and compute capacity is rising I can’t see a world where token pricing will go up.
