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

A zero-instrumentation, continuous CPU–GPU profiler built in Rust using eBPF. It traces CUDA kernel launches, memory transfers, and synchronization points to help identify CPU↔GPU bottlenecks in pr…

Machine Learning Systems

"RAG-Anything: All-in-One RAG Framework"

MLCommons Inference Endpoints repository

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

An acceleration library that supports arbitrary bit-width combinatorial quantization operations

Transforms PDF, Documents and Images into Enriched Structured Data

Function graph tracer for C/C++/Rust/Python

Python tool for converting files and office documents to Markdown.

This curriculum module introduces foundational concepts for solving the Navier-Stokes equations, including methods for interface advection and fluid-structure-interaction.

Fast and memory-efficient exact attention

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

Curated list of project-based tutorials

Linear algebra foundation for the Rust programming language

Parsl - a Python parallel scripting library

An easy-to-use and fast library for task-based parallelism, utilizing coroutines.

BCC - Tools for BPF-based Linux IO analysis, networking, monitoring, and more

Open-source deep-learning framework for building, training, and fine-tuning deep learning models using state-of-the-art Physics-ML methods

Reference implementations of MLPerf® inference benchmarks

Training material on writing machine learning code with PyTorch by ICCS

Physics Informed Deep Learning: Data-driven Solutions and Discovery of Nonlinear Partial Differential Equations

Material for gpu-mode lectures

Slides, notes, and materials for the workshop

Development repository for the Triton language and compiler

CUDA Library Samples