microsoft/generative-ai-for-beginners

21 Lessons, Get Started Building with Generative AI

What is this project?

Generative AI for Beginners is a free, open-source course created by Microsoft Cloud Advocates. It teaches you how to build applications that use generative AI — the technology behind tools like ChatGPT that can create text, images, and more.

The course is a GitHub repository (a public code library) that anyone can access, copy, and learn from.

What does it include?

The course has 21 lessons, each covering a different topic. You can start at any lesson since each one is self-contained.

Lessons come in two types:

  • "Learn" lessons — explain a generative AI concept (like what large language models are, or how to use AI responsibly)
  • "Build" lessons — explain a concept AND include working code examples in Python and TypeScript

Each lesson contains:

  • A short video introduction
  • A written lesson in the README file
  • Code samples that work with either Azure OpenAI or the OpenAI API
  • Links to extra resources for continued learning

What topics are covered?

The lessons span the full range of generative AI development, including:

  • Fundamentals: What generative AI is, how large language models (LLMs) work, and how to compare different models
  • Prompt engineering: How to write effective instructions for AI models
  • Building apps: Text generation, chat applications, image generation, and search apps using vector databases
  • Advanced techniques: Retrieval Augmented Generation (RAG), fine-tuning models, and AI agents
  • Practical concerns: Designing good user experiences, securing AI applications, and managing the AI application lifecycle
  • Specific model families: Small Language Models (SLMs), Mistral models, and Meta models
  • Open-source models: How to use models from Hugging Face

What do you need to use it?

To run the code, you can use any of these:

  • Azure OpenAI Service (Microsoft's cloud AI service)
  • Microsoft Foundry Models (Microsoft's model catalog)
  • OpenAI API (the direct API from OpenAI)
  • Foundry Local — run models fully offline on your own device, no cloud subscription needed

You'll also need:

  • Basic knowledge of Python or TypeScript (the README links to beginner courses for both)
  • A GitHub account to copy (fork) the repository

There's a dedicated Course Setup lesson (lesson 00) that walks you through configuring your development environment.

Who is it for?

It's aimed at beginners — people who want to start building generative AI applications but may not have deep AI experience. The README notes that basic Python or TypeScript knowledge is helpful but not strictly required.

Other notable features

  • Multi-language support: The course is translated into 50+ languages, maintained with a tool called Co-op Translator. Because translations make the repository large, the README includes instructions for downloading only the English version.
  • Community support: There's an official Discord server where learners can connect and get help, plus a developer forum for questions and feedback.
  • Related resources: A separate .NET edition exists for .NET developers, and the team has produced other courses (like LangChain for Beginners).
  • Contributions welcome: The project accepts bug reports and code contributions via GitHub issues and pull requests.

In short

This is a structured, hands-on, 21-lesson course that takes you from knowing nothing about generative AI to being able to build real applications — with code examples, videos, and community support, all free and open source.

Related

  • Project
    rasbt/LLMs-from-scratch

    Official code for the book *Build a Large Language Model (From Scratch)* – a step-by-step, from-scratch implementation of a GPT-like LLM in PyTorch, covering text preprocessing, attention, model implementation, pretraining, and finetuning. Runs on laptops, with optional GPU acceleration.

  • Project
    infiniflow/ragflow

    RAGFlow is an open‑source Retrieval‑Augmented Generation platform that adds agentic reasoning to LLMs. It ingests many document types, creates searchable embeddings, and supports multi‑step retrieval with configurable “thinking” depth. Features include template‑based chunking, knowledge compilation (wikis, graphs, mind‑maps), grounded citations, and plug‑in LLM/embedding models. Deploy locally via Docker or from source; a hosted cloud version is also offered. Licensed under Apache‑2.0 and backed by an active community.

  • Project
    open-webui/open-webui

    Open WebUI is a self-hosted AI platform that provides a user-friendly interface for interacting with AI models, supporting local models via Ollama and any OpenAI-compatible API. It offers features like plugins, role-based access control, persistent memory, RAG, and more, and can run entirely offline.

  • Project
    langgenius/dify

    Dify is an open‑source LLM app development platform with a visual workflow builder, RAG pipeline, autonomous agents, multi‑model support, and built‑in observability. It can be run via Docker‑Compose for self‑hosting or used as a SaaS offering.