Databricks at AWS
re:Invent

Your data is scattered across warehouses, lakes and streaming pipelines. And your AI agents can't reason over what they can't reach. In this session, learn how a single governed data foundation on AWS unifies batch, streaming and operational workloads so agentic AI and real-time analytics run on the same copy of data. See how Lakehouse, Lakebase and real-time pipelines eliminate redundant infrastructure, cut latency and give every agent the context it needs, without moving data. You'll leave with a practical blueprint for consolidating your stack and accelerating time-to-insight. 

Enterprise data becomes useful to AI only when agents can understand the right data, context and permissions. In this session, Databricks will introduce the Genie family, including Genie One, Genie Agents and Genie Ontology. Then S&P Global Energy will show how they made a complex structured data estate conversational using domain-specific Genie Agents, Unity Catalog governance and MCP. Learn how their teams let subject matter experts curate trusted agents, expose them as managed MCP servers, compose them across domains and benchmark answer quality, reducing the time to launch new conversational data experiences from months to days.