NODES AI 2026: Lakehouse + Graph: Delta Lake, Unity Catalog and Neo4j for Governed AI Pipelines

This talk features Shaurya Agrawal at NODES AI, focusing on integrating Delta Lake with Neo4j to construct governed AI pipelines. The session delves into the bronze/silver/gold architecture of Delta Lake, exploring methods for tracking lineage, deriving graph features, and implementing GraphRAG, all while maintaining role-based access control, audit logs, and synchronized data contracts between systems. Attendees will gain insights into architectural patterns that allocate responsibilities for events, aggregations, and feature tables, as well as techniques for lineage tracking, relationship modeling, and syncing Unity Catalog metadata with Neo4j's schema. Participants can expect to leave with reference architecture, code samples, and a design for a governance dashboard that encompasses both lakehouse and graph technologies.