just spent an hour helping a friend debug some lovable-generated code. he's on hour 6 with this bug

super bearish on vibecoding with llms. they're great for prototyping but cannot be trusted

we need a different model architecture that can actually understand and reason about the code

building /cabin, advising /clanker and /farcon | grin.io

I think if you're an excellent developer and can write excellent documentation, from what I hear from devs better than me, you can do a sort of documentation-driven-development approach

But you have to have a fine intuition for the LLM limitations and be prepared for careful code review