Mastodawn

RE: https://neuromatch.social/@jonny/116325668039992121

Hilarious -- and insightful -- analysis of part of the leaked Claude code base. I've been thinking recently about what the code would look like if you treat it as a "black box" and only let an llm touch it. It seems reality gets awfully close to the prediction: "Workaround-based development" and "An even bigger ball of mud."

Show thread

anders 18h ago

Will it get better as models evolve? I don't think so, because it's a systematic problem. Every little inconsistency or other piece of desinformation in a code base will mislead the llm and generate even more noise. It's like thermodynamics. The only way to reduce the noise is by doing work/spending energy on it (a.k.a. refactoring). And that doesn't seem to fit with the "black box" approach.