AI Engineering from Scratch
435 lessons. 20 phases. Write the backprop, the tokenizer, the attention mechanism, and the agent loop by hand before any framework gets imported. Python, TypeScript, Rust, Julia.
FIG_000 · curriculum v1.0 · 2026 open source · MIT AI Engineering from Scratch Every published lesson. Every phase. Every algorithm built from raw math before a single framework gets imported. Maintained by Rohit Ghumare and contributors. Run on your own machine. Your agent becomes your tutor: placement quiz, personalized path, lessons taught interactively in your terminal. FIG_001 · FORWARD PASS y = σ(Wx + b) x h₁ h₂ ŷ Read by engineers and students at “Obsessed with the AI Engineering from Scratch repo.” - AI engineer at Google How this works Most AI material teaches in…
saved by
related reading
- About · AI Engineering from Scratchaiengineeringfromscratch.com
- Building Effective AI Agents \ Anthropicanthropic.com
- Together AI | The AI Native Cloudtogether.ai
- Ichigo (@iiiichigo_chan) on Xx.com
- Harness Engineering for Self-Improvement | Lil'Loglilianweng.github.io
- Breaking into AI Engineering in 2025.newsletter.swirlai.com
- Effective context engineering for AI agents \ Anthropicanthropic.com
- LLM Powered Autonomous Agents | Lil'Loglilianweng.github.io
- Welcome to Learn Harness Engineeringwalkinglabs.github.io
- Latent.Space | Substacklatent.space
- GenAI Handbookgenai-handbook.github.io
- Trending Papers - Hugging Facepaperswithcode.com