Sponsored By: DataHub | Agents That Get It Right: Grounding Databricks Genie in Enterprise Metadata

Answering business questions with AI agents sounds simple — until your agent queries the wrong table, misunderstands what "revenue" means in your organization, or returns a number no one can trace back to a source. The missing ingredient isn't a better model; it's context.In this talk, we'll show how teams are combining Databricks Genie with DataHub, the leading open-source metadata management platform, to build conversational analytics agents that actually work in production. By grounding Genie in DataHub's governed metadata — glossary terms, column-level lineage, dataset ownership, and business context — agents canroute queries correctly, return trusted answers, and cite the data asset behind every response. We'll walk through the architecture, share concrete patterns for connecting Genie to your enterprise metadata layer, and show how teams can validate agent reliability before rolling these out beyond the data team. Attendees will leave with a practical blueprint for turningtheir Databricks investment into an agent that any business user can trust.