Skip to content

How to Use This Hub

This page explains how the AI Agents Learning Hub is structured so you can navigate it effectively and get the most out of every lab.

📊 The Level System

Every lab is tagged with a level from 50 to 400, inspired by Microsoft's conference session numbering (a widely understood signal of depth).

Level Badge Name What to expect Account needed
50 L50 Awareness Reading and concepts. No tools, no account. ❌ None
100 L100 Foundations First hands-on labs. Minimal setup. ✅ GitHub free
200 L200 Intermediate Code + free-tier cloud (GitHub Models). ✅ GitHub free
300 L300 Advanced Cloud services, integration patterns. ✅ Azure subscription
400 L400 Expert Production architecture, evaluations, cost. ✅ Azure paid

Start at the right level

Don't skip L50/L100 even if you're an experienced developer — the conceptual framing helps you make better decisions later.

🗺️ Learning Paths

Labs are grouped into 8 Learning Paths, each focused on a tool or technology:

Path Best for Entry point
🤖 GitHub Copilot GitHub users, developers L100
🏭 Microsoft Foundry Azure developers, ML engineers L200
🔌 MCP Anyone building agent integrations L100
🧠 Semantic Kernel Python / C# developers L100
📚 RAG Developers working with documents/data L100
👥 Agent Builder — Teams M365 developers L100
💻 Agent Builder — VS Code Extension developers L100
⚙️ Pro Code Agents Senior engineers, architects L200

Each path has an index page with the full list of labs, a recommended order, and a short overview of what you'll build.

💡 Suggested Learning Routes

Route A — "Zero to Agent" (No account required to start)

Lab 001 → Lab 002 → Lab 003 → Lab 012 → Lab 020 (Python or C#)
Go from zero knowledge to running your own MCP server locally, no cloud account needed.

Route B — "GitHub-only" (Free GitHub account only)

Lab 010 → Lab 013 → Lab 014 → Lab 022 → Lab 023
Use GitHub Copilot, GitHub Models, and Semantic Kernel — everything free, no credit card.

Route C — "Full Azure Stack"

Lab 013 → Lab 020 → Lab 030 → Lab 031 → Lab 032 → Lab 033
Requires Azure subscription. Build a complete Foundry Agent with MCP, pgvector, RLS, and observability.

Route D — "Teams Developer"

Lab 001 → Lab 011 → Lab 024
Build Copilot Studio agents and Teams bots step by step.

🔖 Reading a Lab Page

Each lab page follows this standard structure:

# Lab XXX: [Title]

[Info box: Level · Path · Time · Cost]

## What You'll Learn
## Introduction
## Prerequisites Setup
## Lab Exercise
  ### Step 1 ...
  ### Step 2 ...
## Summary
## Next Steps
  • Info box at the top tells you everything at a glance: level, estimated time, and cost/account requirements.
  • Prerequisites Setup tells you exactly what to install or configure — with direct links to free trial sign-ups.
  • Lab Exercise is the step-by-step walkthrough.
  • Next Steps links to the natural continuation of the lab.

💰 Free vs. Paid — Our Commitment

We believe the best learning resources are accessible. Our goals:

  • ✅ L50 and L100: Zero cost, zero credit card, just a free GitHub account
  • ✅ L200: Use GitHub Models — free inference, no credit card
  • ⚠️ L300: Azure free tier where possible; clearly marked when an Azure subscription is needed
  • ⚠️ L400: Paid Azure resources required — estimated costs noted in each lab

→ See Prerequisites & Accounts for the full setup guide.