From Raw Documents to AI-Ready Data — Leo Platzer & Jeff Koss

Why watch

This talk addresses the critical challenge of scaling AI document processing from pilot to production, using a manufacturer building a legal-ops chatbot over 80,000 SharePoint files as a case study. It covers AI-assisted taxonomy generation, metadata tagging with evidence and confidence scores, sensitive data detection, and quality checks for duplicates and freshness. The presenters demonstrate how cleaning stale and duplicated data (30% of corpus) doubled recall in multi-hop RAG evaluations, with up to 80% of retrieved context being useless without proper data hygiene. Includes generating context files for coding agents.