Opus reviewing an entire codebase is pretty impressive. I use it a lot for some hobbyist stuff and its great at finding bugs, security issues, and just basic stuff like spelling/grammer mistakes.

But of course, on Lemmy you’ll get downvoted because “ai bad”