Learn JAX

A complete path from JAX fundamentals to production model code. Each track is a self-contained chunk; taken end-to-end they cover the language, the math, the libraries, and the patterns you'll use to ship.

A path is a run of tracks in order. Each one stands alone, so you can start anywhere and come back.

JAX Autodiff

Forward and reverse mode, custom derivatives, gradient checkpointing, per-example gradients. The autodiff toolbox.
JAX Foundations

Functional ML — pure functions, vmap, pytrees. The JAX way of thinking.
JAX Vectorization & Control Flow

vmap, scan, while_loop, fori_loop, cond, switch — JAX's primitives for batching and control flow.
JAX Stochasticity

Distributions, reparameterization, sampling techniques, MCMC, gradient estimators. Randomness as a controllable resource.
JAX Performance & Distributed

jit details, memory and device control, sharding APIs (Mesh, PartitionSpec, shard_map), mixed precision. The performance and parallelism toolbox.
JAX Conceptual Deep-Dives

Custom pytree nodes, host interop callbacks, quantization patterns, jaxpr inspection, advanced debugging. Beyond the curriculum.
JAX Numerical Computing

Linear algebra, decompositions, FFT, convolutions, sparse arrays, polynomial fitting, numerical stability. Scientific computing in JAX.
JAX Profiling & Debugging

Profiler hooks, debug primitives, fixing common tracer/concretization errors, AOT compilation. Practical-skill problems for production work.
Optax

Gradient transforms, optimizer chains, schedules, weight decay, EMA, masking. The production optimizer library for JAX.
Flax

Modules, layers from scratch, attention, transformer architectures, training loops, lifted transforms. Production model code in JAX (Linen API).
Flax NNX

Flax's modern object-oriented API. Stateful modules, eager-mode workflow, split/merge state, lifted transforms, training, sharding. Standalone path through the recommended API for new projects.