# AI Engineering From Scratch

A free curriculum of over five hundred lessons across twenty phases teaching AI engineering hands-on, from math foundations through building agents and MCP servers, in Python, TypeScript, Rust and Julia.

Each lesson ships a reusable artifact, such as a prompt, a skill, an agent or an MCP server, and the README asks a learner to read the lesson, actually type and build the code rather than treat it as decoration, run the command from the repository root, and keep evidence of the exit code and the artifact produced before moving on.

Content is organized goal-first rather than lesson-first: paths exist for someone new to the field, someone who wants math and machine-learning foundations, production LLM engineering, agent engineering, using coding agents on real repositories, product judgment for deciding what to build, the Model Context Protocol, and writing Agent Skills, plus certification preparation tracks. The same lesson codes are used on GitHub and on a companion website.

English is the canonical language, with machine-translated companion pages in several others. MIT licensed.

Website: https://aiengineeringfromscratch.com