The course in one idea

Prompting frontier models is powerful but the abstraction is leaky, and frontier models are closed and cost ~$1B to train. So we can't reproduce the frontier — but we can learn what transfers: the mechanics of how things work and the mindset of squeezing every bit of efficiency from data and hardware. Five parts, five assignments, one recurring question: what is the best model you can build with a fixed data and compute budget?

1 · Basics
tokenize · model · train
2 · Systems
kernels · parallelism · inference
3 · Scaling laws
predict before you spend
4 · Data
eval · curate · clean
5 · Alignment
SFT · DPO · RL

Lectures

Arc: tokenization → architecture → systems → scaling → data → alignment → reasoning → multimodality → inference systems.

Source: CS336 Spring 2026 playlist (Stanford Online) · cs336.stanford.edu. Notes are a rendering of each lecture for quick study.