🏭 Microsoft Foundry Path¶
L200 L300 L400
Microsoft Foundry (Azure AI Foundry) is Microsoft's enterprise-grade platform for building, deploying, and managing AI agents at scale. It brings together model catalog, agent orchestration, evaluation, monitoring, and governance under one unified portal and SDK.
What You'll Build¶
By the end of this path you will have:
- ✅ Navigated the Azure AI Foundry portal and deployed a model
- ✅ Built an AI Agent with tools and MCP integration
- ✅ Connected a PostgreSQL database via MCP for semantic search
- ✅ Secured data with Row Level Security
- ✅ Observed and traced agent calls with Application Insights
- ✅ Evaluated agent quality with the Azure AI Evaluation SDK
Path Labs (7 labs, ~525 min total)¶
| Lab | Title | Level | Cost |
|---|---|---|---|
| Lab 009 | Azure OpenAI Service Quickstart | L100 | ⚠️ Azure |
| Lab 030 | Microsoft Foundry Agent Service + MCP | L300 | Free |
| Lab 032 | Row Level Security for Agents | L300 | ✅ Free |
| Lab 033 | Agent Observability with Application Insights | L300 | ✅ Free |
| Lab 049 | Foundry IQ — Agent Tracing with OpenTelemetry | L300 | ✅ Free |
| Lab 074 | Foundry Agent Service — Production Multi-Agent Deployment | L300 | ✅ Free |
| Lab 050 | Multi-Agent Observability with GenAI Semantic Conventions | L400 | ✅ Free |
Microsoft Foundry Key Components¶
| Component | Description |
|---|---|
| Model Catalog | 1,800+ models including OpenAI, Meta, Mistral, Cohere |
| Agent Service | Managed agent runtime with tool calling, code interpreter, MCP |
| Evaluation SDK | Quality metrics: groundedness, coherence, relevance, safety |
| Azure Monitor | Distributed tracing, metrics, Application Insights integration |
| Content Safety | Built-in responsible AI guardrails |
Prerequisites¶
Azure subscription required
All labs in this path require an Azure subscription.
→ Set up your Azure account