Crystallize is a lightweight Python framework for rigorous, reproducible data science experiments. Define modular pipelines, apply treatments as experimental variations, and verify hypotheses with statistical tests, all with built-in caching, immutability, and parallelism for trustworthy, efficient results.
AI pull-request reviewer: hybrid RAG + a code graph + Claude Code. Whole-repo context, inline GitHub comments grounded on exact code. MCP server + Claude Code plugin.