Before AI, I felt like I could review PRs and, based on whose code I was reviewing, I could be more trusting and casual with my review, since I knew they usually wrote good code and paid attention to details. Now, with AI, no one pays attention to details, so I have to deeply scrutinize every PR, which takes twice or 3x the effort.
AI agents almost always add things that aren’t necessary, or they follow anti-patterns based on the surrounding legacy code context.
@ramsey Yeah, I’ve noticed a distinct drop in quality since coworkers started using LLMs.
It’s hard to be sure, because it’s made me judge every PR more harshly anyway. But I notice some distinct patterns, like specifying default values unnecessarily.
And usually a failure to notice appropriate opportunities for abstraction (because hey, the LLM is so good at boilerplate that we don’t need to worry about that any more, right?)
We had a couple of big PRs of ‘nice-to-have’ features that were too low-priority ever to get done. Vibecoding meant they were initially written in a couple of hours — but then they took literal weeks of back and forth in code review to get them into a tolerable state.
Ben Ramsey
ben
Anders Anni