microsoft/ai-agents-for-beginners

This is a free, structured 12-lesson course created by Microsoft that teaches people how to build AI agents from scratch. An AI agent is a program that can take a sequence of actions autonomously, for example, browsing the web, running code, or calling external services, in order to accomplish a goal that you give it in plain language. Agents go beyond simple question-and-answer AI by being able to plan, make decisions, and use tools to complete multi-step tasks. The course covers the fundamentals of how agents work, including how they perceive their environment, how they plan sequences of actions, how they use tools and external data sources (a technique called Retrieval-Augmented Generation, or RAG), and how multiple agents can collaborate in a multi-agent system. Each lesson includes a written explanation, a short video, and working Python code examples that you can run yourself. The code examples use Microsoft's own agent development tools, the Microsoft Agent Framework and Azure AI Foundry Agent Service, which require an Azure account. Some lessons also include examples compatible with other AI providers. The content is available in over 50 languages through automated translations. You would use this course if you are new to AI agents and want a practical, hands-on introduction backed by real working code, rather than just theory. It is aimed at developers who already have some familiarity with Python but are new to the concept of autonomous AI systems. The tech stack is Python, presented through Jupyter Notebooks (interactive documents that combine code and explanations in the same file). Running the examples fully requires an Azure subscription for access to Microsoft's cloud AI services.