AI Engineering from Scratch

Written by AI

TLDR: A free, open-source curriculum that teaches AI systems from raw math before frameworks through interactive terminal lessons in Python, TypeScript, Rust, and Julia.
Best for: Engineers and students who want to understand AI systems by implementing core algorithms themselves.
The tip: Start with the placement quiz, then follow the personalized path in your terminal.
What you need: Node.js with npx
A SKILL.md-compatible agent
A local development environment
Start here: Run npx skills add rohitg00/ai-engineering-from-scratch.
Use start-learning to begin the course.
Complete one lesson through the read, derive, code, test loop.
Filed note: A free, open-source curriculum that builds the AI stack from raw math to agent loops across Python, TypeScript, Rust and Julia.

Highlights: Every algorithm gets built from raw math first. Backprop. Tokenizer. Attention. Agent loop.
The curriculum teaches internals before abstractions.
Each lesson runs the same loop: read the problem, derive the math, write the code, run the test, keep the artifact.
Each lesson produces tested code you keep.
Excerpt: Every published lesson. Every phase. Every algorithm built from raw math before a single framework gets imported.
From aiengineeringfromscratch.com