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

great tools for shared understanding of a workable mvp. use demo data, anticipate the code is shit and hopefully you can use 30%. you still need to code after prototype.

wireframe -> high fidelity design
vibe proto -> real implementation

The former are great for reducing rework in the latter.