About
Why I Built This
I spent the better part of a year teaching myself how large language models actually work, not the marketing version, the actual math underneath tokenization and attention and gradient descent. I figured a topic this important would already have a mountain of free resources behind it. What I found instead was a scattered mess: incredible papers that assume you already have a PhD, a handful of genuinely good paid courses that assume you have a few hundred dollars to spare, and a lot of surface-level content in between that never actually gets to the math. This site is my attempt to close that gap, for free, for anyone.
I found an interest in machine learning back in college and have built a few small projects with it since. None of that made me an expert, but it convinced me the ideas behind modern AI are a stack of learnable concepts, not magic, and that anyone curious enough can follow the whole thing from derivatives to a working transformer if someone actually explains it clearly.
Everything here, every lesson, every widget, every line of real running code, is free. No account, no credit card, no catch.
One quick note: this is an independent personal project, built and maintained entirely on my own time. It doesn't represent, and shouldn't be taken to represent, the views or interests of my employer.
This project is still growing. A proper way to request missing topics or report confusing bits is coming soon. Until then, find me here: