I just finished writing a Kafka consumer to migrate data with heavy AI help

Yeah, the AI-generated bugs are really insidious. I also pushed a couple subtle bugs in some multi-threaded code I had AI write, because I didn't think it through enough. Reviews and tests don't replace the level of scrutiny something gets when it's hand-written. For now, you have to be really careful with what you let AI write, and make sure any bugs will be low impact since there will probably be more than usual.

> I’ve seen something closer to a 10-20% improvement. The seems to match my experience in "important" work too; a real increase but not essentially changing the essence of software development. Brook's "No Silver Bullet" strikes again...

What about generating testable code? I mean you mentioned detecting subtle bugs in generated code - I too have seen similar - but what if that was found via generated test cases than found by a human reviewers? Of course the test code could have bugs, but I can see a scenario in the future where all we do is review the tests output instead of scrutinising the generated code...

> My real worry is that this is going to make mid level technical tornadoes... Yes! Especially in the consulting world, there's a perception that veterans aren't worth the money because younger engineers get things done faster. I have been the younger engineer scoffing at the veterans, and I have been the veteran desperately trying to get non-technical program managers to understand the nuances of why the quick solution is inadequate. Big tech will probably sort this stuff out faster, but much of the code that processes our financial and medical records gets written by cheap, warm bodies in 6 month contracts. All that was a problem before LLMs. Thankfully I'm no longer at a consulting firm. That world must be hell for security-conscious engineers right now.