What it is
AI for Beginners is a free, self-paced curriculum published by Microsoft on GitHub, written primarily by Dmitry Soshnikov with Jen Looper as editor. It runs across 12 weeks and 24 lessons, organised into sections on symbolic AI, neural networks, computer vision, natural language processing, other AI techniques (including genetic algorithms and multi-agent systems) and AI ethics.
What a learner can do at the end
By working through the lessons and labs, a learner will have covered the basics of neural networks and deep learning, built and trained models with PyTorch and TensorFlow/Keras, used OpenCV for computer vision tasks, and worked with embeddings and transformer-based approaches to NLP. Each lesson includes a pre- and post-lesson quiz and, for most topics, a hands-on lab in a Jupyter notebook.
Format and time commitment
It's entirely self-paced: there are no cohorts or scheduled sessions. Learners fork or clone the GitHub repository and work through lessons in order, running notebooks locally, in Visual Studio Code, or in GitHub Codespaces without any local setup. Microsoft states the course structure spans 12 weeks at a typical study pace, but there's no enforced schedule.
Prerequisites and price
The course is pitched as beginner-friendly with no formal prerequisites stated, though the labs assume basic familiarity with Python. The whole repository is MIT-licensed and free to use, fork or adapt — there is no paid tier, certificate fee or subscription.
Support
Learners can get help through an official Discord server and the Azure AI Foundry Developer Forum linked from the repository, rather than through direct instructor contact.
