I don't know what kind of company has got a large codebase and it's taking 20,000 loc pr's expecting that shit to be sustainable, but I work AI first and at the most PR's are touching 12 to 15 files, with the more radical ones sitting at 30. They are still human-reviewable. They are still the kind of change you can revert back to original without breaking all the rest... there are ways of doing this stuff, dear luddites.

I was wondering how anyone could reach the token limit of a 200$/month Claude subscription. I never did on a 20$ one, and I'm parallelizing Claude to the absolute limit of what I can reasonably keep in mind. And I do that day and night, on week-end too with open-source. Now I'm absolutely convinced that it's by doing crap architectures and design that require 200k token of context before adding a single checkbox. If you apply good design principle and course correct the AI in real time before it commits to really stupid way to fix things (i.e. you don't ask it or let it do things that would make zero sense if you had to do it yourself), then the vibe coding cost way less and is way faster and easier to reason about.

I’m hardly a Luddite and you’re basically making my point.

You’re deliberately constraining the output so changes stay small, human-reviewable and reversible. That’s good engineering culture.

It isn’t all or nothing. I use AI heavily but that doesn’t mean I have to pretend there aren’t serious problems with how it’s being used.