Microsoft Foundry

This course contains the use of artificial intelligence.

Unlock the full power of Microsoft Foundry and learn to build autonomous AI agents that reason, search, remember, and collaborate. This comprehensive, hands-on course takes you from core Azure AI services all the way to production-grade multi-agent systems — with real, executable Python code in every lesson.

What You'll Build:

You won't just watch — you'll build. Starting with individual AI capabilities like language detection, text translation, speech synthesis, document extraction, and content moderation, you'll progressively assemble these building blocks into increasingly sophisticated intelligent agents. By the end, you'll architect multi-agent workflows where specialized agents collaborate autonomously to solve complex real-world problems.

What's Inside:

Foundry Tools — Hands-on with six core Azure AI services: Language, Translator, Speech, Search, Document Intelligence, and Content Safety. Each demo includes Infrastructure-as-Code (ARM templates) so you can deploy every resource with a single command — no portal clicking required.

Building Your First Agent — Create a conversational AI agent in Azure AI Foundry, give it instructions, connect it to conversations, and watch it respond intelligently.

Agent Memory & Context — Build agents that remember user preferences across sessions using memory stores and embedding models for semantic search.

Code Interpreter — Equip your agent with the ability to write and execute Python code on the fly — performing data analysis, generating charts, and processing CSV files in a sandboxed environment.

Connecting Agents to the World — Integrate external REST APIs via OpenAPI specifications, connect to MCP (Model Context Protocol) servers, add real-time web search, and combine multiple tools in a single agent that intelligently decides which tool to use.

Multi-Agent Workflows — Orchestrate complex systems using the Microsoft Agent Framework. Build sequential pipelines where one agent's output feeds into the next, parallel workflows where agents work simultaneously, and cloud-hosted agents with real-time streaming responses.

Production Monitoring with KQL — Monitor your agents in production using Kusto Query Language and Azure Application Insights. Track token usage, trace agent reasoning, analyze errors, and build operational dashboards.

Infrastructure as Code — Demos includes ARM templates with detailed comments, parameter files, and deployment instructions. Deploy Azure resources in the correct configuration state without touching the Azure Portal.

This course contains promotional materials.