PAIRS (PArallel Images for eveRyday Scenarios)

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

Efficiently Composable Data Augmentation on the GPU with Jax

KFAC from scratch (KFS)---Paper & Code

A post-modern modal text editor.

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

Flax is a neural network library for JAX that is designed for flexibility.

Adaptive Experimentation Platform

Laplace approximations in JAX.

Unbearably fast near-real-time pure-Python runtime-static type-checker.

Library for reading and processing ML training data.

Type annotations and runtime checking for shape and dtype of JAX/NumPy/PyTorch/etc. arrays. https://docs.kidger.site/jaxtyping/

Tensors and Dynamic neural networks in Python with strong GPU acceleration

Solutions of Reinforcement Learning, An Introduction

The Incredible PyTorch: a curated list of tutorials, papers, projects, communities and more relating to PyTorch.