Turn APIs, tools, and data into real agent velocity | DEM331
Chu Lahlou demonstrates how Azure AI Foundry can turn existing enterprise APIs, tools, and data into callable capabilities for AI agents, with grounding via Foundry IQ and runtime governance to keep tool usage observable and controlled.
Overview
This Build 2026 demo focuses on enabling AI agents to securely use real enterprise capabilities (APIs, tools, and internal data) rather than operating only on generic model knowledge.
Key themes called out in the session description:
- Using Foundry to let agents call real APIs and use tools securely.
- Using Foundry IQ to provide grounding and context from enterprise data.
- Applying runtime governance so tool usage is observable and controlled.
Session chapters (from the video description)
- Agents struggle with reading and interpreting messy data
- Demonstration of audio file analysis for crew inspections
- Using CU Studio to access prebuilt models
- Defining analyzers for multiple document types
- Demonstration of sending documents through analyzer pipeline
- Results from root cause and material planning workflow
- Introduction to implementing with Agent Framework
- Follow-up question using cached results in Agent Framework
- Resources, GitHub code, and upcoming breakout session announcement