Had a chat with a dev yesterday that flipped my take on AI code assistants

He told me he spends more time fixing AI-generated bugs than just writing the code himself. I laughed at first, but then he showed me 4 errors in a 30-line function that took him 20 minutes to untangle. Anyone else finding these tools create more work than they save in the long run?

4 comments

4 Comments

the_daniel1mo ago

Nah @adam186, pretty sure the real wall is devs who just don't know how to prompt right.

adam1861mo ago

Honestly, I think everyone's missing the real issue here. These tools are trained on average code from the internet, so they spit out mediocre solutions by default. They're great for boilerplate or when you're stuck on syntax, but the second you need anything clever or specific to your project's weird edge cases, they fall apart. The devs who swear by them probably just haven't hit that wall yet or they're writing really simple stuff. Ngl, it's like trusting a GPS that only knows the main roads and then blaming you when it sends you down a dead end.

wade_anderson1mo ago

Mark had it recommend some database library that didn't even exist yet.

henry_anderson543d ago

Gotta disagree here. I've seen plenty of devs get great results on complex stuff by just being more specific with their prompts and explaining the weird edge cases upfront.