Adi Ganesh

Speculative decoding for LLM inference

When two language models are faster than one.

My local LaTeX workflow

Live-compilation, macros, and more.

GPT in words and code

Notes on transformers / LLMs from the ground up.

The modern algorithmic toolbox

One of my favorite technical classes at Stanford.

The principle of maximum entropy

Introducing entropy and its applications in statistics / ML.

Pólya-Burnside enumeration in combinatorics

Using group theory for combinatorics.

The Meissel-Lehmer algorithm

Deriving an algorithm to efficiently compute `\pi(x)`.