Hacker News new | past | comments | ask | show | jobs | submit login

I agree - it’s hard to enter a flow state while reviewing someone else’s - or some AI’s - code. That’s a major reason why I haven’t started using LLMs for code, personally.

I am glad someone else feels this way. Maybe it’s not going to be as big a paradigm shift as I originally expected.




It will probably raise the floor a lot. The least competent (not meaning in a bad way! I was one of those) coders will be a lot more competent all of a sudden.


I'm skeptical. It's easy to make something sound correct at a first glance but that has subtle fundamental flaws that invalidate it.

Knowing humans and the LGTM phenomenon, these kinds of issues will slip by quite readily.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: