Open-Source AI Engineering Course: 523 Lessons from Linear Algebra to Autonomous Agents

The open-source project 'AI Engineering from Scratch' (52.6K stars) offers a comprehensive curriculum: 20 stages, 523 lessons, ~342 hours. It covers foundations (math, ML, deep learning), perception & language (CV, NLP, speech, transformers, generative AI, RL), LLMs (training from scratch, engineering, multimodal), agents (tools, MCP, agent engineering, multi-agent systems), and production & responsibility (infrastructure, ethics, capstone). Each lesson follows a six-beat structure: Motto → Problem → Concept → Build It → Use It → Ship It. Every lesson produces a reusable artifact (prompt, SKILL.md, agent definition, MCP server). The course is AI-native: can be taught by coding agents via npx skills add.