BRK297 Breakout Tue 17 Nov 16:30 PST
The shift from RAG to context: Grounding agents with Foundry IQ
Pablo Castro
Summary
From the recording / Microsoft AI summary
The session covered easier knowledge base creation inside Foundry, including ingesting, chunking, vectorising and connecting content. Foundry IQ Serverless is due to reach general availability in December for production retrieval.
Session description
Microsoft's catalogue description
The market is moving from app-specific RAG to shared context layers to give many agents secure access to trusted enterprise knowledge. We'll explore how Foundry IQ (Azure AI Search) improves grounding across the stack: from core search innovations and knowledge graphs to purpose-built models that make agentic retrieval faster, higher quality, and easier to configure. We'll also show how Foundry IQ fits into Microsoft IQ, enabling people and agents with easier knowledge discovery and consumption.