Build the data layer for enterprise AI | THRSP1064
Microsoft Events presents an intermediate-level session on connecting enterprise data and business context for AI. The walkthrough includes OneLake, Fabric IQ, and Eon, with demonstrations of querying and governing data across clouds.
Connecting distributed data to enterprise AI
Enterprise data often resides across SQL Server, Azure SQL, Blob Storage, and other systems. The session examines how to make that data accessible to Copilot, AI agents, and Power BI while avoiding additional pipelines, compute, storage, and data movement.
The approach combines unified data access with continuously maintained business context:
- OneLake provides unified enterprise data access and governance.
- Catalog federation and shortcuts connect distributed data.
- Apache Iceberg interoperability supports access through an open table format.
- Semantic models and ontologies give AI agents business context.
- Fabric IQ brings enterprise data and context together for AI use.
Integration, context, and governance
The session covers zero-copy, multi-cloud integration between Eon and OneLake. It also introduces Eon’s autonomous data foundation, including backup-driven Apache Iceberg ingestion and data classification.
The live demonstration brings these elements together through zero-ETL integration, ontology agents, and AI-powered querying across clouds. Security topics include automated sensitive-data classification, OneLake role integration, governance, and protection of personally identifiable information (PII).
Session chapters
- 00:00 — Introduction to building the data layer for enterprise AI
- 00:48 — Challenges of answering enterprise AI questions across distributed, multi-cloud data
- 02:36 — OneLake overview: unified enterprise data access and governance
- 03:51 — Catalog federation, shortcuts, and Apache Iceberg interoperability
- 05:26 — Zero-copy, multi-cloud integration between Eon and OneLake
- 07:01 — Semantic models and ontologies for AI agent business context
- 10:18 — Eon’s autonomous data foundation, backup-driven Apache Iceberg ingestion, and classification
- 14:05 — Automated security, sensitive-data classification, and OneLake role integration
- 15:42 — Live demonstration of zero-ETL integration, ontology agents, and AI-powered cross-cloud querying
- 19:26 — Automated governance, PII protection, and closing remarks
Resources
Session: THRSP1064 · Cloud & AI · Theater · Intermediate (200) · Microsoft Ignite 2026