Loading…
25
At a small dev meetup in Tacoma last month, a guy showed me his AI code review bot and now I can't decide if I love it or hate it
I was at this tiny meetup in Tacoma last month, maybe 20 people in a back room of a brewery, and this guy named Dev sits down next to me and shows me his AI code review bot. He runs it on every pull request at his job and it flags stuff like variable names and missing tests, which honestly sounds fine at first. But then he says he stopped reading his teammates' code altogether, the bot does it and he just hits approve if the green check shows up. Half the room thought that was genius, saving like 6 hours a week, and the other half looked like someone farted. I get both sides. The bot caught a real bug for him once, an off-by-one in a loop that would've shipped to production. But it also told him a perfectly fine function was too long, and he split it into three worse functions just to make the warning go away. I keep going back and forth on whether this stuff is a tool or a crutch, and I genuinely don't know. Where do you all draw the line with AI reviews, do you still read the actual diff or just trust the bot?
0 comments
Log in to join the discussion
Log In0 Comments
No comments yet
Be the first to share your thoughts on this discussion.