Agentic AI With Open Source Data Architecture

This talk presents a comprehensive framework for developing intelligent autonomous systems using an integrated open source data architecture. I demonstrate how Apache Iceberg, Polaris, and SQLMesh create a unified data foundation that addresses the key challenges in building effective Agentic AI. This architecture provides transactional consistency, schema evolution, and semantic understanding while eliminating data duplication and synchronization issues. Built on this foundation, Agentic AI systems can analyze information, make decisions, and take actions with minimal human supervision. Real-world applications across customer service and data management demonstrate significant efficiency gains while maintaining governance and security. The approach ensures flexibility, reduces costs, and enables AI systems to leverage trusted data through a consistent interface—creating a scalable path toward increasingly autonomous intelligent systems.​​​​​​​​​​​​​​​​