Course focus: Microsoft's Generative AI for Beginners course contains 21 lessons that alternate between concept-focused learning and application-building exercises.
What you will learn: Learners move from LLM selection and responsible AI through prompts, text and chat apps, vector search, image generation, function calling, UX, security, RAG, open models, agents, and small models.
Recommended order: Complete the course setup lesson, choose Python or TypeScript, and use one provider path consistently. Alternate Learn and Build lessons, storing every working example in a private practice repository.
Before you begin: Basic Python or TypeScript is helpful. Running examples requires access to Microsoft Foundry Models, Azure OpenAI, or the OpenAI API; provider availability and billing are separate from the free course.
Who it suits: Useful for beginners who know basic programming and want a structured path from concepts to small generative-AI applications.
Important limits: The repository notes that GitHub Models is retiring at the end of July 2026, so new learners should follow the maintained provider paths in the current setup lesson rather than older screenshots. Use environment variables or local secret files excluded from Git, set provider budgets, and remove personal data from prompts and sample datasets.
Project and license: Repository content is provided under MIT; check the current LICENSE before reuse. Repository: https://github.com/microsoft/generative-ai-for-beginners
Break the goal into input, action, and output before adding review checkpoints.
Keep examples, failure conditions, and review criteria so the tutorial stays reusable.