Pointless rant. Please ignore. I’m a software developer and we all know how AI has changed our industry. How we work or why we’re fired and why we can’t afford PCs.
Anyways, we’re already all forced to use AI already and we’re already atrophying the minds of our juniors. It’s great.
New team meeting and one of our managers tells us that we’re never going to write code anymore at all. The AI will read the JIRA ticket and create the pull request (change request to the codebase) on GitHub. Our job is to only review the code on GitHub and then rank how well AI did and then comment and then get AI to fix it. We have to do this so we can improve the AI process. Which is funny because none of the people who plan this AI shit are data scientists. The only way they can change things is by promoting, it’s not like we’re releasing our own coding models but anyways … He’s like, now you should be able to do much more work and just review PRs all day now and that we should never be doing only one thing. You can only tell AI through a GitHub comment to fix a mistake and then you can start reviewing the next thing.
We were like, if it’s a simple fix why can’t we just fix it?
“Because we need to improve the AI process”
But then, I have to context switch.
“Yes that’s the point you can come back to it later”
Why come back to it later when we can solve it now? We can even use AI to solve it now.
“No, we want you just comment on the PR so the bot can handle it”
Context switching is free apparently… It’s actually infuriating because apparently we’re not using IDEs any more. I personally use the GitHub plugin to review PRs in my IDE but no one else seems to do it so I don’t think they even took that into account.
These guys have auto merged AI code that’s taken us weeks to unravel and which we still haven’t fully been able to fix. They just merge shit all the time and a lot of it is fucking slip. AI merged hundreds of tests and no one cares when they break. They didn’t configure prettier because AI doesn’t use it so it breaks out formatting when humans do it.
I ranted to my own manager for 30 minutes about it today and he was just as upset because every developer is now asking what exactly are they doing. My manager asked me what I would do. I said the process sucks but what are we supposed to do as devs. If I review 20 PRs a day, how is the company going to ensure my skills are gonna be sharp? What are we doing about taking in ideas from regular devs? How do we ensure code ownership when we’re just merging tickets we don’t write and code we had no hand in shaping?
Sorry. I actually thought I had faith in my company with AI because they were coming up with thoughtful approaches but it seems like utter incompetence.
Run like fifty agents in parallel and get lauded for how “productive” you are while you find another job and hopefully before they get the bill.
We have to do this so we can improve the AI process.
…To be clear, does this manager think y’all are “improving” the model with feedback? Like it will learn or train off your comments?
Or what? Are they just talking about the intermediate steps to the LLM API?
I’m just trying to wrap my head around this.
Most of the things that try to address this do so by adding information to the context window. Doesn’t solve the fundamental issues, like LLMs still being solely dependent on word/token correlations and the assumption that if you throw enough conversations at it and do statistical analysis on the words used, you can encode the specific knowledge behind those words used into those statistics and then models can get back from those generalized statistics to discuss specific topics (which isn’t really how statistics work).
Well that’s often counterproductive, too. Even the largest models degrade if you fill their context with stuff they don’t need, or push them outside the patterns they were finetuned for.
Not that RAG isn’t really important.
Back when it was just “add a note about the specific thing in the hidden prompt”, it felt like an old-style chat bot was patched on top of the LLMs, where it had a series of specific responses to watch out for and fix. Now it feels more like there’s a large database of this shit and we just hope that the LLM will look up the correct stuff when it needs it (and probably prompt it to do that when demoing it for others and creating impressive examples that it probably won’t live up to very well in the real world unless each user knows how to direct it with follow up prompts when it inevitably goes in the wrong direction).
I don’t know because these guys aren’t data scientists … they’re devs so they can only do this by prompting so it’s just gonna be whatever steps they add to the agentic workflow.
By “these guys” you mean a separate team setting up an agentic workflow just for your workplace?
I’ve finetuned some models for a job, not just LLMs, and I can all but guarantee they aren’t finetuning some self hosted coding model. The APIs they use do not “learn” from mistakes, so if that’s what management is thinking, they are misinformed.
And interesting things like constrained output are often not even supported by APIs.
Modern LLM APIs and wrappers come with really good tool harnesses for coding, too. Like, good enough you don’t want to screw with it and pollute the context structure the original trainers tuned it for.
What’s more… not only do models change, but single model’s capabilities change under load, for more speed during peak usage times. So there is no guarantee some configuration they find that “works better” will work better later. See:
What I’m saying is… I wonder how long this mess last?
Because if you already have GH PR integration, most of what an agentic team can do is mess things up.
I’m in the same position as you. The Devs job is now to write specfiles which Claude implements. It’s been months since I wrote code myself and I’m already forgetting basics. This industry is fucked.
This happened at our shop fifteen years ago. We were instructed to outsource everything to offshore.
That didn’t work at all, because with a few rare , bright exceptions the people the offshore company have us could only achieve an outcome if they had a list of steps for that specific thing.
It’s going to be exactly the same thing with vibe coding. It kind of works in the hands of somebody with a deep understanding of the tech, but they expect to hand it to juniors and get good result.
So we’ll either have to pick up the pieces or let them flounder.
Here’s the question that worries me the most: Across how many critical industries is this happening, and at what scale?
It feels to me like decision makers are pushing for the creation of mountains of code no one fully understands and potentially creating massive issues with systems we can’t afford to have go down. Are we creating decades’ worth of technical debt in a matter of months and eroding systems that our societies need to function because people just want to save money and be part of the fad? How bad is this going to get?
I see this as a Darwinian force to weed out many companies. Which I think is good in today’s environment.
Smaller, fitter organizations , or even individuals, will have a chance to shape technology
Imagine when companies like GE, Safran, RR, start letting AI slop into their FADECs and other avionics
In case it wasn’t already clear to you, they’re having you train the AI so it can literally take your job and they can lay you off.
Oh no they don’t, they think they’re having them train the AI to lay them off, but it’s absolutely not what’s actually happening.
You can’t just train the AI. Not unless you own some of the more sophisticated RAG-based solutions that tell the AI how they should behave. Or if you’re producing your own LLMs. But in both cases - the process is vastly different if you want it to actually work (for RAG, you have to actually write some consistent knowledge base and build a MCP server around it). Now they’re just chatting with the robot.
Like - buddy, your coffee machine ain’t gonna get better the more times you click the espress button.
Source: I’m engineering my own kb framework for AI to know how to write code EXACTLY like I do. And I need that only because I want to focus on architecture and devops now that I mastered and burned out of fullstack.
You mastered fullstack?
South Park needs to revisit and update that episode. Dey Turk er jerbs!
I’m having early 2000s flashbacks. Overseas developers were going to replace us all for a fraction of the cost. We had to turn over our code…almost like training AI. That lasted until the first SQL server update that broke, literally, everything they did. Then they wanted a huge amount of money and weeks to get it working again. That really put management in a bind because they told us not to touch their code. Ooookay, fine. Meanwhile, I’m looking at 2,000 character URL string containing the database admin account and password in plain text.
If I were you, I would burn soooo many tokens. This is a malicious compliance dream scenario.
Fuck the owner class, get paid, and watch it all burn.
Opus large 1m context. /Plan and thinking mode. Use cli. Converse with it Think deeply. Think step by step. Read the entire codebase every turn. Bloat context. Wait. Sharpen your skills and wait for the accountants to be the new heroes
Unionize and collectively bargain. You owners are trying to get you to teach your replacement. Right now you are 1 squeaky gear in the great machine of your corporation. The c-suite isn’t bothered by it. But if every gear and cog on the dev team worked together you could end it.
I just saw a skit on youtube about a company stealing keyboards so the employees have to talk to AI chatbot to code. Hold on, lemme find it… Found it: https://youtu.be/eBRzdfU-K30
Knew it would be AlbertaTech before I opened it lol
The github comment thing is insane lmao.
I know its the classic dumb suggestion, but best to clean up your resume just in case.
Should make a complaint that github commenting is too slow and that devs need direct access via an API/dev key so that they can use AI to speed up the PR review process significantly.
Then watch the ensuing flames as they make a pipeline that automatically reviews every PR using the same LLM lol.
/s?
I just went on a 3-week burner vibe coding shit. Everything was garbage, and it barely works. But I had an impossible deadline.
I’m now going back to manually coding most things. Using AI to debug or make changes. Which is especially helpful when there’s changes to be made across a large codebase.
You could silently sabotage the system. Approve code with obvious defects, bad code smell, and vulnerabilities. Let the product go to production, crash, and let the company spend a bunch of money refactoring the horseshit vibecoded stack they doubled down on.
Im sorry youre going through this. It’s fucking frustrating. Just get that money, fuck the system, and find joy elsewhere
Oof. I’ve had bad jobs, but being told not to do your job so you can instead directly help AI replace you!? I would absolutely not do this, or at least not plan on being there for very long…
or at least not plan on being there for very long…
Especially since maintainence will become a nightmare before long
I am here definitively seeing an uptick of job ads for senior software developers “with good debugging skills”.
You know what they are for.
Unfortunately, I bet HR will still filter out tons of them for bullshit excuses so it takes a year to fill out positions
So my coworker and I discuss this all the time and we figure we’re old enough to just not give two shits, follow the process, pick up a FOSS project or two to keep our skills up, and wait for the inevitable of when they figure out AI slop is killing them and then charge outrageous prices as contractors when AI2K hits.
AI2K
Ooh, I’m gonna try to use this one
Better yet, clone a big repository like Linux kernel or Firefox and let it crawl all day while you’re still working there. Let them pay for the tokens and win the tokenmaxxing game.
Remember, when they call you asking to return you can gouge the price the fuck up. Beauty of capitalism
Engineering is a joke here.
Our manager doesn’t care. I’ve got some coworkers who are very pro-AI though. Every time we talk to them, they mention the word “AI” every five minutes, talking about how we shall let AI write our code, try out the new “agent” and “skill” they’ve made.
For the love of god, those “agents” are simply fucking markdown file! You wrote a crap document which isn’t even grammar-checked!
People channeling code reviews written by AI to GitHub’s PR comment section and call it a day. Damn it. I asked you to read my code and give your idea both between lines and from higher up, and you give me some shit I could have generated by myself! Your credibility has dropped to zero.
And what is up with writing test cases with AI? Oh, just because it doesn’t fuck up the production even if it becomes shit? Guess what, fuckers, it does! Test cases are an important line of defense against bugs. It ensures no bad behaviour is introduced after a change. We’d better trust a strong security dude guarding this gate, no? You fuckers what, put fucking Frankenstein at that position? I trust the code you wrote, not some fucking probabilistic machine slop! Who cares how good the code AI generates? It was you who I trusted, not “it”! You broken the chain of trust and I guess I won’t be trusting you from now on!
For the love of god, those “agents” are simply fucking markdown file! You wrote a crap document which isn’t even grammar-checked!
I ran something on Opus 4.8 or something two weeks ago and it ignored my rules and when asked why it was like, “yeah I read it but i ignored it”
That’s another thing about this is when they say: “tell us what PRs are bad so we can improve the process”
None of these guys are data scientists … what are they going to do? Bloat the context? There’s reasons for why any PR isn’t acceptable and any instruction they’re secretly writing can change over time.
