Tutorials you can run
Long, interactive explanations of how machine learning actually works. Nothing is hidden behind a library, every number is visible, and all of it runs in your browser — including the Python.
Each one is a single subject taken slowly, in parts. Read them in order or land in the middle; every part says what it needs on its own.
- 10 parts Neural networks, explained by building one Invent an alphabet that has never existed, draw it by hand, and train a neural network to read your handwriting — in your browser, with every number visible at every step.
- 12 parts A tiny GPT, built from scratch Start by counting which letter follows which, and end with a transformer. Every piece hand-written before any library version is allowed near it.
What these are
Every tutorial here is built the same way: you make something real in the first few minutes, and every term is demonstrated before it is named. There is no video, nothing to install, and nothing to sign up for in order to read.
Anything you draw stays in your own browser. Which pages get read is counted, and nothing else: no cookie, no identifier, nothing kept once the tab is closed. There is no record of you to build, which is why there is no banner asking your permission to build one.
They are free, and they stay free. If something here teaches you the wrong thing, that is a bug — report it.