TL;DR
An open-source curriculum of 503 lessons covering the entire AI engineering process from math basics to autonomous agents, learned through hands-on implementation.
Key features
Step-by-step learning path with 20 phases and 503 lessons
Implement all algorithms directly in four languages: Python, TypeScript, Rust, Julia
Each lesson produces reusable artifacts such as prompts, skills, agents, and MCP servers
Systematic structure: Math (linear algebra) → ML basics → Deep learning → Vision/NLP/Speech → Transformers → LLM → Agents/Production
MIT license, completely free, runs in local environment
When to use it
Developers who want to systematically learn AI engineering from scratch
Students or juniors aiming to grow into AI engineers with both theory and practice
Educators or teams needing a comprehensive curriculum to supplement fragmented AI education