Today I had a chat with a colleague and he told me that I shouldn't delegate to AI any code that AI it's just there to help you but not to generate the code for you. Not sure on what to think here, for one side I get what he is saying, basically you end up loosing control of what…
Somebody said if you delegate the code to the AI, AI owns the code. You won't be able to do any change in a reasonable time without the AI. To be honest I don't really care as long database schema and migrations are architected and designed by me. The same with the code…
What I'm more worried about is losing my "muscle memory" when it comes to both writing code and the problem solving it involves. I've already started to feel that, and I've been using AI since the beta of Github CoPilot.
I have no problem with AI generating most of the code. But I'd like to think "the more code AI generates, the more I have to review" and vice versa.
Most of the time, the agent ususally write the same kind of code that I would, but twice (if not more) as fast.