HW1 is also public! Ever wondered how to build an agent harness from scratch? Now you can! With less than $5 of API credits, you can build your own "Claude Code" at home! Repo: (2/5)
Open-source AI evals toolkit: skills that guide AI coding agents to write eval rubrics, eval datasets (User Input Grid), trace codes, code evaluators and LLM-as-a-judge prompts. From the Reforge AI Evals course by Calibre.