Running large language models on a single GPU for throughput-oriented scenarios.

A thread-safe queue faster and more resource efficient than golang's native channels

[NeurIPS'23] H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models.

Extremely Fast End-to-End Deep Multi-Agent Reinforcement Learning Framework on a GPU (JMLR 2022)

atomate2 is a library of computational materials science workflows

[ICML 2025 Spotlight] ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference

quacc is a flexible platform for computational materials science and quantum chemistry that is built for the big data era.

Open-source red teaming framework for MLLMs with 42+ attack methods

Realtime distributed messaging platform built using Go and React (Fullstack) ❤️

One database for user-user (follows), user-item (likes), and item-item (related items) interactions — precomputed at write time, served as simple lookups

High-performance scalable message broker for modern event-driven applications

⚡ Zero-Stall MoE Inference via Lookahead Prediction & Async DMA Prefetching. Optimized for SSD I/O with Hybrid MLA+Sliding Window Attention.

A snakemake-based workflow for FEP and MM(PB/GB)SA calculations with GROMACS

extremly light uart library for AVR 8 bit microcontrollers

A high-performance, SIMD-optimized, pure C library for reading and writing Apache Parquet files.

A low-latency LRU approximation cache in C++ using CLOCK second-chance algorithm. Multi level cache too. Up to 2.5 billion lookups per second.

Aperod APRO - Institutional-grade Layer-1 blockchain. RingCT/MLSAG zero-knowledge privacy engine with 1-second deterministic BFT finality for high-throughput dApps.

An Intuitive, Lightweight, High Performance Full Stack Java Web Framework.

🔬🖥 Automated Blood Vasculature Analysis of 3D Light-Sheet Image Volumes

A reactive driver for Aeron transport (https://github.com/real-logic/aeron)