agentic analytics over 100M+ rows

naive RAG wasn't going to cut it, so i built an agent that writes and runs its own queries.

this one started in the RFP phase with the IDB: a chatbot over survey datasets in the hundreds of millions of rows. naive RAG falls over at that scale, so i built an agent-orchestration system with a code-execution tool that queries the database and runs Python analysis on demand. delivered, and the bank's researchers praised it right alongside the document chatbot.