William Zhang

xla. A machine learning compiler for GPUs, CPUs, and ML accelerators

fix-exchange. C++ equity exchange with FIX order entry, a price-time matching engine, UDP market data broadcast, and SQLite logging/crash recovery

chisel. Chisel: A Modern Hardware Design Language

OpenDRAMmodelV1. Open DRAM Model v1.0 (Access Transistors & High-Voltage Periphery Transistors for 6F²/4F²/3D DRAM)

OpenTPU. A open source reimplementation of Google's Tensor Processing Unit (TPU).

chipyard. An Agile RISC-V SoC Design Framework with in-order cores, out-of-order cores, accelerators, and more

mlir-beginner-friendly-tutorial. This is a beginner-friendly tutorial on MLIR from the perspective of a user of MLIR, not a compiler engineer. This tutorial will introduce why MLIR exists and how it is used to compile code at different levels of abstraction. This tutorial will focus on working with the "core" dialects of MLIR.

macro-place-challenge-2026. Macro placement engine combining electrostatic global placement, Lin-Kernighan k-opt, and LAHC

cmux. Open source Ghostty-based macOS terminal with vertical tabs and notifications for AI coding agents. Built for multitasking, organization, and programmability.

marimo. A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Python. All in a modern, AI-native editor.

circt. Circuit IR Compilers and Tools

NVmain. NVMain - An Architectural Level Main Memory Simulator for Emerging Non-Volatile Memories

DRAMSys. DRAMSys a SystemC TLM-2.0 based DRAM simulator.

ramulator. A Fast and Extensible DRAM Simulator, with built-in support for modeling many different DRAM technologies including DDRx, LPDDRx, GDDRx, WIOx, HBMx, and various academic proposals. Described in the IEEE CAL 2015 paper by Kim et al. at http://users.ece.cmu.edu/~omutlu/pub/ramulator_dram_simulator-ieee-cal15.pdf

skills. Public repository for Agent Skills

cli-agent-orchestrator. Python

awesome-copilot. Community-contributed instructions, agents, skills, and configurations to help you make the most of GitHub Copilot.

Xplace. Xplace 3.0: An Extremely Fast, Extensible and Deterministic Placement Framework with Detailed-Routability and Timing Optimization

Fuser. A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")

FlashMLA. FlashMLA: Efficient Multi-head Latent Attention Kernels

Open-L2O. Open-L2O: A Comprehensive and Reproducible Benchmark for Learning to Optimize Algorithms

TileKernels. A kernel library written in tilelang

AutoDMP. C++

polar_quant. PolarQuant: Fused KV cache decode kernel that ties cuBLAS on B200 with 4.1x compression

SOL-ExecBench. A benchmark of real-world DL kernel problems

tvm-ffi. Open ABI and FFI for Machine Learning Systems

macro-place-challenge-2026. Python

linear_algebra. Solutions to Linear Algebra Done Right, by Sheldon Axler.

COSMO.jl. COSMO: Accelerated ADMM-based solver for convex conic optimisation problems (LP, QP, SOCP, SDP, ExpCP, PowCP). Automatic chordal decomposition of sparse semidefinite programs.

littlefs. A little fail-safe filesystem designed for microcontrollers

tilus. Tilus is a tile-level kernel programming language with explicit control over shared memory and registers.

casadi. CasADi is a symbolic framework for numeric optimization implementing automatic differentiation in forward and reverse modes on sparse matrix-valued computational graphs. It supports self-contained C-code generation and interfaces state-of-the-art codes such as SUNDIALS, IPOPT etc. It can be used from C++, Python or Matlab/Octave.

RLBench. A large-scale benchmark and learning environment.

microMLC. minimal compiler

CircuitNet. CircuitNet: An Open-Source Dataset for Machine Learning Applications in Electronic Design Automation (EDA)

AIOS. AIOS: AI Agent Operating System

tiny-gpu. A minimal GPU design in Verilog to learn how GPUs work from the ground up

riscv-sodor. educational microarchitectures for risc-v isa

c-compiler. A C Compiler in Rust

MLSys. Hosts MLSys Competition Track Problems

claudes-c-compiler. Claude Opus 4.6 wrote a dependency-free C compiler in Rust, with backends targeting x86 (64- and 32-bit), ARM, and RISC-V, capable of compiling a booting Linux kernel.

openclaw. Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

snmalloc. Message passing based allocator

grokking-the-object-oriented-design-interview. Python

intel-graphics-compiler. C++

hlsl-specs. HLSL Specifications

DirectXMath. DirectXMath is an all inline SIMD C++ linear algebra library for use in games and graphics apps

sparse-llm.c. LLM training in simple, raw C/CUDA

bullet3. Bullet Physics SDK: real-time collision detection and multi-physics simulation for VR, games, visual effects, robotics, machine learning etc.

jax-js. JAX in JavaScript – ML library for the web, running on WebGPU & Wasm

mlir-tutorial. a comprehensive, step-by-step guide to creating a new MLIR dialect outside the LLVM source tree.

cvdp_benchmark. Python

original_performance_takehome. Anthropic's original performance take-home, now open for you to try!

Quant-Developers-Resources. Resources to Prepare for Quant Developers/ Quantitative Researcher/ Quantitative Trader/ Quant Analyst/ Software Engineers in Quant Trading Firms, HFTs and Hedge Funds

lerobot. 🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

cvxpygen. Code generation with CVXPY

hardcaml_xilinx_components. Hardcaml Xilinx component definitions

llama2.c. Inference Llama 2 in one file of pure C

mirage. Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

KernelBench. KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)

jax-gs. Python

flow_matching. A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.

DiT. Official PyTorch Implementation of "Scalable Diffusion Models with Transformers"

Vitis-HLS-Introductory-Examples. ReScript

nextpnr. nextpnr portable FPGA place and route tool

BitNet. Official inference framework for 1-bit LLMs

cocotb-cpp. This repository explores writing cocotb-style tests in modern C++, using coroutines and strong typing, with the goal of maintaining a Python-like test style while potentially improving simulation performance.

NeuSim. An open-source simulator framework for neural processing units

torch-rnn. Efficient, reusable RNNs and LSTMs for torch

densecap. Dense image captioning in Torch

micrograd. A tiny scalar-valued autograd engine and a neural net library on top of it with PyTorch-like API

libmoon. LibMOON is a standard and flexible framework to study gradient-based multiobjective optimization.

PyTorch-Quantization-Aware-Training. PyTorch Quantization Aware Training Example

Two-Layer-Hierarchical-Softmax-PyTorch. Two-Layer Hierarchical Softmax Implementation for PyTorch

rerun. Visualize, query, and stream to train on multimodal robotics data.

tinyml. Python

ateenysitp. a whirlwind tour to deep learning and deep learning systems

Stock-Trader. A trading bot that uses convex optimization

ecos-python. Python interface for ECOS

CUDALibrarySamples. CUDA Library Samples

taco. The Tensor Algebra Compiler (taco) computes sparse tensor expressions on CPUs and GPUs

torchtitan. A PyTorch native platform for training generative AI models

circuit_sparsity. Open-source release accompanying Gao et al. 2025

cuda-tile. CUDA Tile IR is an MLIR-based intermediate representation and compiler infrastructure for CUDA kernel optimization, focusing on tile-based computation patterns and optimizations targeting NVIDIA tensor core units.

sentence-transformers. State-of-the-Art Embeddings, Retrieval, and Reranking

hardcaml_waveterm. A terminal based digital waveform viewer for Hardcaml

tinytinyTPU. a mini 2x2 systolic array and PE demo

dgSPARSE-Lib. PyTorch-Based Fast and Efficient Processing for Various Machine Learning Applications with Diverse Sparsity

asst5-kernels. Python

perfetto. Production-grade client-side tracing, profiling, and analysis for complex software systems.

gemmini-rocc-tests. Fork of seldridge/rocket-rocc-examples with tests for a systolic array based matmul accelerator

gemmini. Berkeley's Spatial Array Generator

zigzag. HW Architecture-Mapping Design Space Exploration Framework for Deep Learning Accelerators

timeloop. Timeloop performs modeling, mapping and code-generation for tensor algebra workloads on various accelerator architectures.

tt-micropython-firmware. TinyTapeout demo pcb's RP2040 functionality

uv. An extremely fast Python package and project manager, written in Rust.

unsloth. Unsloth is a local UI for training and running Kimi K3, Gemma 4, Qwen3.6, DeepSeek-V4, GLM and other models.

kvrocks. Apache Kvrocks is a distributed key value NoSQL database that uses RocksDB as storage engine and is compatible with Redis protocol.

leetgpu-challenges. LeetGPU Challenges

cutile-python. cuTile is a programming model for writing parallel kernels for NVIDIA GPUs

tt-ece298a-demoscene. Verilog

GeneticAlgorithmPython. Source code of PyGAD, a Python 3 library for building the genetic algorithm and training machine learning algorithms (Keras & PyTorch).

llm-council. LLM Council works together to answer your hardest questions

coralnpu. A machine learning accelerator core designed for energy-efficient AI at the edge.

exafmm-t. A kernel-independent fast multipole method library with Python interface.

DFFRAM. Standard Cell Library based Memory Compiler using FF/Latch cells

caravel. Caravel is a standard SoC template with on chip resources to control and read/write operations from a user-dedicated space.

breakout-pcb. Tiny Tapeout breakout PCBs

gf180mcu-pdk. PDK for GlobalFoundries' 180nm MCU bulk process technology (GF180MCU).

lbvh. an implementation of parallel linear BVH (LBVH) on GPU

torq-compiler. Torq compiler sources

circuit_training. Python

ml-systems-course. minimal DL library in C: 24 NAIVE cuda/cpu ops, autodiff engine, python API (ops bindings/layers/models), tensor abstraction, strides, complex indexing (multi-dim slices like numpy), computation-graph visualizer, saving/loading params, primitive garbage collection, autogenerated tests, optimizers, weight initializers, etc. (no AI assistance used)

Systolic_MAC_with_DFT. GF180 ASIC tapeout of a 2x2 MAC with DFT infrastructure

walrus. Python

presidio. An open-source framework for detecting, redacting, masking, and anonymizing sensitive data (PII) across text, images, and structured data. Supports NLP, pattern matching, and customizable pipelines.

TPUv2. a mini TPU with floating point arithmetic

sam3. The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

carbon-lang. Carbon Language's main repository: documents, design, implementation, and related tools. (NOTE: Carbon Language is experimental; see README)

tilelang. Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

vtr-verilog-to-routing. Verilog to Routing -- Open Source CAD Flow for FPGA Research

swig. SWIG is a software development tool that connects programs written in C and C++ with a variety of high-level programming languages.

modular. The Modular Platform (includes MAX & Mojo)

nqcc2. Reference implementation for Writing a C Compiler

nqcc. A compiler for a tiny (but growing!) subset of C, written in OCaml.

SWAGGER. SWAGGER: Sparse WAypoint Graph Generation for Efficient Routing

hardcaml_verify. Hardcaml Verification Tools

hardcaml. Hardcaml is an OCaml library for designing hardware.

rmm. RAPIDS Memory Manager

cuvs. cuVS - a library for vector search and clustering on the GPU

nwm. Official code for the CVPR 2025 paper "Navigation World Models".

PLDM. Python

jepa. PyTorch code and models for V-JEPA self-supervised learning from video.

dino_wm. Python

faiss. A library for efficient similarity search and clustering of dense vectors.

omnilingual-asr. Omnilingual ASR Open-Source Multilingual SpeechRecognition for 1600+ Languages

accelerated-computing-hub. NVIDIA curated collection of educational resources related to general purpose GPU programming.

llvm-project. The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.

transformers. 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

CLIP. CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

linux. Linux kernel source tree

executorch. On-device AI across mobile, embedded and edge for PyTorch

stablehlo. Backward compatible ML compute opset inspired by HLO/MHLO

iree. A retargetable MLIR-based machine learning compiler and runtime toolkit.

asap7_pdk_r1p7. Perl

xls. XLS: Accelerated HW Synthesis

XNNPACK. High-efficiency floating-point neural network inference operators for mobile, server, and Web

sglang. SGLang is a high-performance serving framework for large language models and multimodal models.

taskflow. A General-purpose Task-parallel Programming System in C++

libfork. A bleeding-edge, lock-free, wait-free, continuation-stealing tasking library built on C++20's coroutines

Task-Execution-Engine. Implementing a thread pool to run with a hardware-optimized number of workers

parrot. Parrot is an array fusion GPU library built on NVIDIA's CCCL libaries (Thrust/CUB).

gluten. Gluten is a middle layer responsible for offloading JVM-based SQL engines' execution to native engines.

BigDL. BigDL: Distributed TensorFlow, Keras and PyTorch on Apache Spark/Flink & Ray

OpenROAD. OpenROAD's unified application implementing an RTL-to-GDS Flow. Documentation at https://openroad.readthedocs.io/en/latest/

folly. An open-source C++ library developed and used at Facebook.

Parallel-QR-Decomposition. Course-grained parallel thin-QR decomposition algorithms for tall-and-skinny matrices on a CPU in C++ using OpenMP.

cvxpylayers. Differentiable convex optimization layers

torch-mlir. The Torch-MLIR project aims to provide first class support from the PyTorch ecosystem to the MLIR ecosystem.

hmmlearn. Hidden Markov Models in Python, with scikit-learn like API

STL. MSVC's implementation of the C++ Standard Library.

eigen. Eigen is a C++ template library for linear algebra: matrices, vectors, numerical solvers, and related algorithms.

botorch. Bayesian optimization in PyTorch

github-contribution-graph-action. 🙈 This GitHub action will automatically push empty commits to one of your repositories.

nanoGPT. The simplest, fastest repository for training/finetuning medium-sized GPTs.

optuna. A hyperparameter optimization framework

GEM. Open-source RTL logic simulator with CUDA acceleration

alpaca-py. The Official Python SDK for Alpaca API

LightGBM. A fast, distributed, high performance gradient boosting (GBT, GBDT, GBRT, GBM or MART) framework based on decision tree algorithms, used for ranking, classification and many other machine learning tasks.

pyuvm. The UVM written in Python

reverie. An ergonomic and safe syscall interception framework for Linux.

community. Kubernetes Community Documentation

iris. AMD RAD's multi-GPU Triton-based framework for seamless multi-GPU programming

ThunderKittens. Tile primitives for speedy kernels

dynamo. A Datacenter Scale Distributed Inference Serving Framework

mlx. MLX: An array framework for Apple silicon

opentitan. OpenTitan: Open source silicon root of trust

ECE298A-TPU. A custom AI chip to be taped out soon!

Optim.jl. Optimization functions for Julia

cvxpy. A Python-embedded modeling language for convex optimization problems.

Triton-Puzzles. Puzzles for learning Triton

DataStructure. Data Structure

CppCoreGuidelines. The C++ Core Guidelines are a set of tried-and-true guidelines, rules, and best practices about coding in C++

OpenSTA. OpenSTA engine

OpenSTA. C++

iris. A neurosymbolic framework for vulnerability detection in code

cuopt. GPU accelerated decision optimization

tpu. yeeeeeeeeeeeehaaaaaaaaaaaaaaaw

tt07-tinytpu. TinyTPU module with only 1 Core and 2x2 matrix multiplication support !

cupynumeric. NumPy and SciPy on Multi-Node Multi-GPU systems

legion. The Legion Parallel Programming System

transformer.cpp. TransformerCPP is a minimal C++ machine learning library with autograd and tensor ops, inspired by PyTorch. It includes a from-scratch Transformer model demo, optimized for CPU and multithreaded performance.

Modern-CPP-Programming. Modern C++ Programming Course (C++03/11/14/17/20/23/26)

JuMP.jl. Modeling language for Mathematical Optimization (linear, mixed-integer, conic, semidefinite, nonlinear)

Design-Patterns. Practicing software design patterns in C++

stable-baselines3. PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.

mujoco. Multi-Joint dynamics with Contact. A general purpose physics simulator.

ModelingToolkit.jl. An acausal modeling framework for automatically parallelized scientific machine learning (SciML) in Julia. A computer algebra system for integrated symbolics for physics-informed machine learning and automated transformations of differential equations

julia. The Julia Programming Language

minbpe. Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.