News for hackers. Stuff that matters.

danggggg

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @08:16AM (#49360509)
Obviously he would have shared them already if they aligned with his abusive rhetoric.
Re:AI usage patterns in software teams on Wednesday August 19, 2026 @08:16AM (#49360503)
I really think -10x is accurate if we're considering the systemic effort it takes to manage the torrent of slop produced by AI addicts and true believers. These AI developers are cancerous to our organizations and codebases.
Re:AI usage patterns in software teams on Wednesday August 19, 2026 @08:13AM (#49360474)
Just do it.

AI chatbots, like social media, condition you to become addicted to short-timescale intermittent reinforcement.

You need to break out of the dopamine spell and do something hard.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @08:11AM (#49360466)
> Even McKinseys and BCGs have a hard time showing ROIs.

That's a pretty low bar if you ask me.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @07:08AM (#49359891)
I'm sorry that you've been tricked. Give up the ghost before you ruin your reputation more than you already have.
Re:AI usage patterns in software teams on Wednesday August 19, 2026 @07:05AM (#49359877)
> vim/emacs-fu is more enjoyable than prompting and waiting.

I couldn't agree more. These AI users go on and on about "tooling" but would never take a week to learn how to use vim. These people are addicts. Don't take their words at face value. They are dependent on AI code generation.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:59AM (#49359826)
> > Usage does not correlate with valued output or ROI

>This sounds possible but how can we know?

If usage correlated with ROI, we would never stop hearing about it from the big AI firms.

The fact is that a bunch of marks and rubes got suckered into buying snake oil.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:50AM (#49359740)
Your bio says that you're a CTO. If you spent millions on AI contracts, you're a mark and you don't deserve your job.
Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:28AM (#49359552)
> I have below average prompt frequency because I know crazy shit like #!/bin/bash and ps aux. It's so goddamn insane.

I think AI made everyone so dumb that I'm a 10x engineer now.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:25AM (#49359530)
Frankly, it seems obvious to me that AI has negative efficiency (0.25x as efficient as the old way? 0.125x as efficient as the old way? 0.0625x as efficient as the old way?)

AI has created a generation of 0.125x engineers.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:24AM (#49359511)
Do you have a rebuttal or are you just mad that people don't like AI?

It sucks at generating code and AI true believers have made my job hell.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:22AM (#49359496)
> I spend 45 minutes writing it. then i git commit and move on cuz i made something good.

I cannot express strongly enough how much I'd rather work with you than with someone writing a 1000-word essay on how AI code generation is actually good enough this time.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:21AM (#49359485)
I have never found AI code generators to be capable of generating code following existing patterns (and really moreso have been disappointed about AI code generators' capabilities in this regard compared to Emacs/LSP tooling).

I could see it being true in a very regimented design, but I'm still skeptical, I'd like to see details of that design, and I question whether very strict adherence to design patterns limits the ability of the LLM to generate useful features.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:17AM (#49359454)
> It's a ludicrous antipattern because person Y still needs to review the PR that person A generated for codebase X, and it will take them about as much time to wrangle the 2000 line PR (oh boy do LLMs love their mocks for unit tests) as it would have been for them to do the 50 line code change.

Do you have any users? Is the software important? I'd find a new job if I were you

> On top of that, it has "allowed" us to add feature after feature onto codebases not designed for them without refactoring. Is it good that this Flask API went from a purpose built service that interacted with the data analytics for product A stored in database X, and now our sales guys can sell product B, C, and D stored in database X? Uh, I'm sure it's great for them. Oh, and now it all can be stored in database X, Y, or Z depending on what the customer wants or what sales promised them. Great. Now I'm looking at a Flask app.py that's 12,000 lines of repeated code.

This is terrifying to me. I'm sorry.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:15AM (#49359427)
> Yeah it’s great for prompting for improvements, granted you can actually judge the suggestions for yourself. They’re often slightly or majorly bad.

I agree with this comment.

That being said, the major errors in some of the suggestions make me terrified of the people who say things like "AI is great for researching and strategizing".

I don't think these people are on the whole engaged or knowledgeable enough to understand the dead ends that AI research is leading them to or the concepts that it doesn't (or can't) discuss.

Really I just feel like AI has turned subpar engineers into dangerous, supercharged Dunning-Kruger cases. Everyone backing AI code generation online acts like a "10x developer", and I think the point of the AI coding tools are to make you feel that way.

I think most of you would be better off slowing down and writing the code by hand in a focused, methodical manner.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:10AM (#49359390)
> So there’s a risk that our judgement will decay over time.

It's happening in real time and those of us who don't use AI code generators can tell.

Re:AI usage patterns in software teams on Wednesday August 19, 2026 @06:08AM (#49359379)
> I think we should think of it less like some intelligent being or “agent” and more as a code generation tool.

I think the jury is still out on whether AI code generation is worthwhile at all. Maybe it's a done deal for management because it allows them to spy on their workers, but I think it's less efficient and significantly more dull than writing code by hand.

Re:AI usage patterns in software teams on Tuesday August 18, 2026 @08:40PM (#49355041)
If you don't accept my opinion as your own you're going to lose your job and become a member of a permanent underclass.
Re:AI usage patterns in software teams on Tuesday August 18, 2026 @08:31PM (#49354976)
> I think it’s near time we all stop having such strong opinions about the matter either way personally.

I don't think this is reasonable given how abusive the pro-AI rhetoric has been for years now.

Re:AI usage patterns in software teams on Tuesday August 18, 2026 @08:17PM (#49354855)
Why not just write the code yourself? To me it seems like methodically constructing the pull request by hand is probably faster than lazily prompting and re-prompting.
Re:AI usage patterns in software teams on Tuesday August 18, 2026 @08:13PM (#49354806)
I was joking. The workflow doesn't seem particularly fast or engaging in my opinion. AI code generation doesn't seem worthwhile to me.
Re:AI usage patterns in software teams on Tuesday August 18, 2026 @08:06PM (#49354725)
That's gotta be at least 10x or 20x more efficient than the old way of doing things.