Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Hypothesis Sage: Agentic RAG Statistics
Learn how to build a CLI tool that queries a statistical knowledge base, selects appropriate tests, generates examples, and chains actions using agentic RAG.
I used to work as a data scientist, and would frequently run into issues understanding and interpreting statistical tests in the real world. Hypothesis Sage is a thought experiment at solving this problem. The tool helps you query a database of statistical info, identify appropriate tests given a situation, create examples of tests given a specific test, and even chain together a set of these tools to answer arbitrary questions. The kicker is that each tool returns information in a parsed, structured format, making it easy for a statistician/data scientist to use this info to proceed in their data analysis.