SQL is becoming a powerful foundation for agentic applications. In this session, we will look at how to prepare relational data for semantic search using external model embeddings, text chunking, and DiskANN through SQL Server’s vector index capabilities. Then we will show how SQL MCP Server gives agents, including agents built in Microsoft Foundry, governed access to that data through security boundaries, permissions, and practical guardrails.

You will leave with a practical understanding of how SQL Server can support modern AI application patterns while keeping enterprise data close, queryable, protected, and safe for agentic workloads.

1. Understand how to prepare relational data for semantic search using external model embeddings, text chunking, and SQL Server vector indexes.
2. Learn how SQL MCP Server enables agents, including agents in Microsoft Foundry, to access SQL data through governed and secure interfaces.
3. Explore practical guardrails for agentic workloads, including permissions, query boundaries, data protection, and database compute safety.