News for hackers. Stuff that matters.
Re:AI usage generates a fragile codebase by nprateem on Wednesday November 11, 2026 @12:51AM (#50450122)
Ok and do you run a few dedicated rounds to improve quality and refactor, add tests and docs etc?

Staying inline with the codebase is the right thing to do. It's down to you to tell it when it's time to re-engineer.

Re:AI usage generates a fragile codebase by nomel on Wednesday November 11, 2026 @01:59AM (#50450466)
> It's down to you to tell it when it's time to re-engineer.

> without excessive, and very often forgotten, instructions to do so.

It forgets, within long sessions. I'm still seeing Opus 5.5 implement workarounds for bugs it created within its own session, even with explicit, and plentiful, instructions and global memories that it's time to re-engineer.

But to your point, yes, multiple rounds are required. My point is that this is going to eventually hold development back.

Re:AI usage generates a fragile codebase by nprateem on Wednesday November 11, 2026 @06:19AM (#50452089)
Just don't use long sessions and always start a new one to check for bugs/divergence before you commit.