—
Finished of the full curriculum spine
20 phases · 473 lessons
AI Engineering from Scratch
Build every algorithm from raw math before importing a framework. Backprop, tokenizer, attention, and the agent loop — written by hand, then verified against production libraries.
—
4
How this works
Most AI material teaches in scattered pieces. A paper here, a fine-tuning post there, a flashy agent demo somewhere else. The pieces rarely line up.
This curriculum is the spine. Linear algebra at one end, autonomous swarms at the other. Every algorithm gets built from raw math first. By the time PyTorch shows up, you already know what it’s doing under the hood.
Each lesson runs the same loop: read the problem, derive the math, write the code, run the test, keep the artifact. No paywall. Fully available on this server.
Progress
Current signal
Live counts from the mirrored curriculum on this host — no external calls.
Finished lessons
0 / 0
Phases
0 / 0
Languages
Python · TS · Rust · Julia
Glossary
···
Table of contents
Twenty phases. One spine.
Tap a phase card to expand its lessons. Each one ships when its math, code, and test are written.
Complete
In progress
Planned
Colophon
Self-hosted curriculum. Lessons, quizzes, and artifacts are mirrored onto this server — open a lesson and it loads locally. No signup. Learn at your own pace.
Open Catalog → pick a lesson → read on this host