M
Most AI material is scattered
A paper here, a fine-tuning post there, a flashy agent demo somewhere else. The pieces rarely line up.
416 lessons across 20 phases. Every algorithm built from raw math before a single framework gets imported. Free, open source, and meant to run on your own laptop — at your own pace.
Open Source. Open Learning.
A paper here, a fine-tuning post there, a flashy agent demo somewhere else. The pieces rarely line up.
20 phases, 416 lessons, four languages. Linear algebra at one end, autonomous swarms at the other.
Read the problem, derive the math, write the code, run the test, keep the artifact. That's it.
tap a phase to peek inside ↓
pythonic clarity
typed in the wild
fast & honest
math, native
The entire curriculum lives on GitHub. Every lesson ships with runnable code in Python, TypeScript, Rust, or Julia — whichever fits the concept best.
$ git clone https://github.com/rohitg00/ai-engineering-from-scratch $ cd ai-engineering-from-scratch $ cd phase-00-setup $ cat README.md # Welcome, builder. # Phase 0 · Lesson 01 — Your workshop. # Read. Derive. Code. Test. Keep.