Using AI agents correctly is a process of reviewing code. If you’re good at reviewing code, you’ll be good at using tools like Claude Code, Codex, or the…
Yeah but here’s the thing: I don’t mind reviewing competent code done by a human being who knows what they’re doing for honest-to-goodness mistakes. But I don’t want to review slop barfed out by a statistical completion machine anymore than I want to review code spewed out by a junior engineer who hasn’t got a clue what the code he copy/pasted does.
I didn’t spend 4 decades busting my balls in this industry to be good at what I do for that. Fuck that.
Yeah but here’s the thing: I don’t mind reviewing competent code done by a human being who knows what they’re doing for honest-to-goodness mistakes. But I don’t want to review slop barfed out by a statistical completion machine anymore than I want to review code spewed out by a junior engineer who hasn’t got a clue what the code he copy/pasted does.
I didn’t spend 4 decades busting my balls in this industry to be good at what I do for that. Fuck that.