Building Code First AI Agents with Azure AI Agent Service
summary
Hands-on demo shows a stateful Azure agent built with the Azure AI Foundry Python SDK that generates SQL, queries SQLite sales data, and merges results with product PDFs stored in Azure AI Search and Storage to produce tables and charts. The agent uses Azure AI Agent Service plus Azure OpenAI for LLMs, a vector index for document grounding, and a sandboxed code interpreter that emits Python to render images. Demos work end‑to‑end but are nondeterministic, struggle with precise math, and need explicit stop/criteria for looping autonomy.
AI-written summary · about
chapters (21)
- 0:00Introduction to AI agents
- 1:51Workshop use case overview
- 3:15Defining AI agents
- 6:33Azure AI Agent Service benefits
- 9:40Agent service workflow steps
- 13:40SQL query generation demo
- 20:47Lab environment setup
- 24:11Agent complexity levels
- 30:00Function calling explained
- 33:14Agent tool specialization
- 40:06Code interpreter setup
- 43:22Agent configuration walkthrough
- 50:53Main application code overview
- 1:00:09SQL query generation demo
- 1:06:53Multi-tool agent demonstration
- 1:18:02Vector database grounding
- 1:24:44Document retrieval demo
- 1:32:34Code interpreter for charts
- 1:39:24Impact analysis demonstration
- 1:48:34Bing search grounding
- 1:51:08Q&A session
related talks
36mBuilding AI Agents in .NET with Azure AI Foundry
56mBuilding AI Agents and Multi-Agent Systems with Azure AI
39mEvolving Intelligence: Designing Autonomous AI Agents
46mAI Agents for the Rest of Us
07h 25mAI Agents Conference 2026 - Day 2
05h 17mBuilding AI Applications
15mBuilding Enterprise AI Agents on Azure
25mBuilding AI Agents