@mattpocockuk curious if you used any tools to solve & compare CRAP level like fallow and does the skill /code-review support that?
it's hard to kept agent automate maintainable codebase and sometimes i'm thinking a model harness choice at a level like flash still cannot achieve that haha
Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.