microsoft /
AI-For-Beginners
A free 12-week, 24-lesson course on AI with notebooks, quizzes and labs
- 70.1k stars
- 769 stars since 3 Oct
- Jupyter Notebook
- Updated yesterday
- Beginner friendly
- Learning & resource lists
This is a self-paced course, not a software library. It runs 12 weeks across 24 lessons. Each lesson has written material, and many have a notebook and a lab to practise on.
It starts with the older symbolic approach, where knowledge is written down as rules and ontologies. It then moves to neural networks and deep learning, with code in TensorFlow and PyTorch. Later lessons cover architectures for images and text, plus less common topics such as genetic algorithms and multi-agent systems.
The README is clear about what it leaves out. Classic machine learning is covered in Microsoft's separate Machine Learning for Beginners course. Business use of AI, Azure cloud services, chatbots and the deep maths are pointed to other resources. It also says its coverage of recent models may lag the state of the art.
What it does
Structured 12-week path
24 lessons come with quizzes and labs alongside the reading.
Notebooks in two frameworks
Lessons include Jupyter notebooks in PyTorch, Keras and TensorFlow, so you can compare them.
Symbolic AI to neural networks
Covers expert systems and ontologies, then perceptrons, multi-layer networks and building your own small framework before using standard ones.
Computer vision track
Starts with OpenCV and moves on to convolutional neural networks and their architectures.
50+ translations
The README lists more than 50 languages and explains a sparse checkout that skips translations to shrink the download.
What you could build
- Follow a full beginner course on AI over about three months.
- Run the Jupyter notebooks to see how a perceptron and a neural network work in code.
- Compare the same neural network written in PyTorch, Keras and TensorFlow.
- Teach an introductory AI class using the lessons, labs and quizzes.
- Learn the basics of computer vision with OpenCV and convolutional networks.
Get started
Clone the repository, using the sparse checkout if you want to skip the translations, then open the course setup lesson to prepare your environment. The README also links a Binder badge for running notebooks online.
git clone --filter=blob:none --sparse https://github.com/microsoft/AI-For-Beginners.gitcd AI-For-Beginnersgit sparse-checkout set --no-cone '/*' '!translations' '!translated_images'Commands are copied from the project's README. Read it on GitHub for the full setup.
More in Learning & resource lists
LLMs-from-scratch
Code for the book that walks you through building a GPT-style model in PyTorch
ai-engineering-from-scratch
Free 20-phase course that teaches AI engineering by building each piece by hand
system_prompts_leaks
A library of system prompts from popular AI chatbots and coding tools
dive-into-llms
Free Chinese notebooks and slides for learning to fine-tune, prompt and secure LLMs
awesome-claude-code
A curated list of guides, skills, tools and plugins for Claude Code users
OpenMAIC
Turns a topic or document into an interactive classroom taught by AI agents
12-factor-agents
A written guide to twelve principles for building reliable LLM-powered software
easy-vibe
A free, multilingual course that teaches AI-assisted coding by building real products
awesome-claude-skills
A curated list of Claude Skills, grouped by task, with links to each one
reasoning-from-scratch
Code for a book that adds reasoning to a small Qwen3 model in PyTorch