Building AI Agents from Zero to Production

🌐 Multi-Language Support

Supported via GitHub Action (Automated & Always Up-to-Date)

Prefer to Clone Locally?

This repository includes 50+ language translations which significantly increases the download size. To clone without translations, use sparse checkout:

Bash / macOS / Linux:
git clone --filter=blob:none --sparse https://github.com/microsoft/Building-AI-Agents-From-Zero-To-Production.git
cd Building-AI-Agents-From-Zero-To-Production
git sparse-checkout set --no-cone '/*' '!translations' '!translated_images'
CMD (Windows):
git clone --filter=blob:none --sparse https://github.com/microsoft/Building-AI-Agents-From-Zero-To-Production.git
cd Building-AI-Agents-From-Zero-To-Production
git sparse-checkout set --no-cone "/*" "!translations" "!translated_images"
This gives you everything you need to complete the course with a much faster download.

A course teaching you the fundamentals of the AI Agent Development Lifecycle

🌱 Getting Started

This course has lessons covering the fundamentals of building and deploying AI Agents.

Each lesson builds on the previous one, so we recommend starting from the beginning and working your way through to the end.

If you want to explore more about AI Agent topics, you can check out the AI Agents For Beginners Course.

Meet Other Learners, Get Your Questions Answered

If you get stuck or have any questions about building AI Agents, join our dedicated Discord Channel in the Microsoft Foundry Discord.

What You Need

Each Lesson has its own code sample that you can run locally. You can fork this repo to create your own copy.

This course currently uses the following:

Microsoft Agent Framework (MAF)
Microsoft Foundry — a project with a deployed GPT-5 series model (for example gpt-5.1). Do not use the retired GPT-4o / GPT-4.1 models.
Azure OpenAI Service
Azure CLI — sign in with az login before running any sample
Python 3.12 or later

Please ensure you have access to these services before getting started.

💰 Cost & cleanup. The hands-on lessons create real Azure resources — a Microsoft Foundry project, a model deployment, a vector store, and (in Lessons 4–6) hosted agents and toolboxes. These can incur cost while they exist. When you finish a lesson — or the course — delete the resources you no longer need. The simplest approach is to put everything in a dedicated resource group and delete the whole group when you're done:
az group delete --name <your-resource-group> --yes --no-wait
You can also delete individual agents, vector stores, and toolboxes from the Foundry portal.

Note on models. This course serves all models through Microsoft Foundry. It does not use GitHub Models, which is being retired on July 30, 2026 — Microsoft Foundry is the official migration path. If you have older code that calls GitHub Models, point it at a Foundry model deployment instead.

More options around model hosting and services coming soon.

🗃️ Lessons

Lesson	Description
Agent Design	An introduction to our "Developer Onboarding" Agent Use Case and how to design effective agents
Agent Development	Using the Microsoft Agent Framework (MAF), build a set of specialised agents to help new developers onboard.
Agent Evaluations	Using Microsoft Foundry, find out how well our AI Agents are performing and how to improve them.
Agent Deployment	Using Microsoft Foundry hosted agents and OpenAI ChatKit, see how to deploy an AI Agent into production.
Production Hosted Agents	Take a hosted agent to enterprise production: Hosted Agents vs Capability Hosts, bring-your-own storage, memory, and governance.
Microsoft Toolbox	Define tools once and govern them centrally: build a toolbox, consume it from an agent via one MCP endpoint, and version tools safely.
Multi-Agent & A2A	Compose agents as networked services: expose an agent over the open Agent-to-Agent (A2A) protocol and consume a remote agent as a peer.

🎒 Other Courses

Our team produces other courses! Check out:

LangChain

Azure / Edge / MCP / Agents

Generative AI Series

Core Learning

Copilot Series

Contributing

This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.

When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.

This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.

Trademarks

This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos is subject to those third-parties' policies.

Getting Help

If you get stuck or have any questions about building AI apps, join:

If you have product feedback or errors while building visit:

Building-AI-Agents-From-Zero-To-Production

Links

README