A machine learning framework for automating bug triage, quality management, and test selection to improve software engineering efficiency.

A React component library for visualizing AI agent traces, turning complex JSON logs into hierarchical visual diagrams for easier debugging.

An open-source engine for Claude Code that turns goals into verified code via graded PRDs, dependency-ordered task graphs, and evidence-based execution.

A single-file relational spreadsheet format based on SQLite, designed for both human users and AI agents to organize and query structured data.

A high-performance Julia framework for agent-based modeling that enables the simulation of autonomous agents reacting to their environment and each other.

NetKet is a Python library built on JAX that uses neural networks and machine learning to study many-body quantum systems.

A syntax-aware code navigation tool that provides structural insights for humans and AI agents, reducing the number of tool calls needed to explore codebases.

A Python library that simplifies the creation of configurable and scalable workflows by eliminating hand-written YAML configs in Hydra projects.

An operating layer for Claude Code and OpenAI Codex that preserves repository state and coordinates multi-step workflows for AI coding agents.

An MCP-native browser control plane that gives AI agents a real browser with human-in-the-loop takeover, reusable auth profiles, and built-in safety rails.

A comprehensive repository of structured AI conference paper records and a local search tool for analyzing academic literature from major venues like NeurIPS and ICLR.

A configurable backend for web, mobile, AI, and IoT apps that provides APIs, authentication, and RAG capabilities as configuration instead of code.

An executable, Git-native catalog for tabular data work that turns ephemeral agent-generated scripts into reproducible, content-addressed pipelines.

A lightweight agent workstation and command center for Codex CLI, Claude Code, and ACP-compatible agents, featuring task scheduling and a unified project workspace.

A tool that analyzes repositories to find and apply the cheapest, most reliable AI coding configuration by testing candidates against real-world tasks.

ParoQuant is an INT4 quantization framework for LLMs that uses learned pairwise rotations to reduce weight outliers and maintain high accuracy during efficient inference.

MInference accelerates the pre-filling stage of long-context LLMs by leveraging dynamic sparse attention patterns to achieve up to 10x speedup on a single A100 GPU.

A pipeline parallel training script that allows for the training of large-scale image and video diffusion models across multiple GPUs.

A unified multimodal backend for AI data apps that combines a database, orchestration, and serving layers into a single Python file.

A comprehensive multimodal dataset and benchmark of 8,150 high-fidelity CFD car designs used to train AI models for aerodynamic optimization and performance prediction.

A no-code GUI and framework for fine-tuning large language models, enabling users to experiment with hyperparameters and advanced training techniques without coding.

A columnar file format by Meta designed for large, wide datasets, optimizing storage and access for machine learning training tables and feature engineering.

PyKEEN is a Python package for training and evaluating knowledge graph embedding models, providing a comprehensive library of built-in datasets and models.

A fast, header-only C++ machine learning library that provides a wide array of ML methods and functions with bindings for Python, Julia, Go, and R.

A hardware-accelerated JavaScript library for training and deploying machine learning models in the browser and Node.js.

An MLOps/LLMOps scheduler and orchestration tool that automates the execution of ML experiments across local and cloud resources with zero configuration.

A multithreaded web crawler that converts website pages into markdown files to simplify document parsing and chunking for RAG and LLM fine-tuning.

Moss is a sub-10 ms semantic search runtime that embeds retrieval and embeddings directly into an application's process to eliminate network latency for real-time AI agents.

Vearch is a cloud-native distributed vector database that enables efficient similarity search of embedding vectors for AI applications.

A local-first Agent OS that lets you run official AI agent CLIs on your own machine and control them remotely via a browser or messaging apps.

A zero-code, visual drag-and-drop big data integration platform that uses AI to help users build data pipelines and query templates without writing code.

A Python library for Difference-in-Differences causal inference providing a wide range of econometric estimators for measuring campaign lift, policy impact, and staggered rollouts.

A Windows desktop application built with Tauri and Rust to monitor DeepSeek API account balances, monthly spending, and token usage trends.

A fast, language-agnostic code navigator that indexes codebases into SQLite to provide millisecond-latency symbol lookups, impact analysis, and structured navigation for humans and AI agents.

An AI-native data notebook platform that replaces Jupyter's JSON format with a human-readable YAML structure for better collaboration and version control.

A Python toolkit for decision making under uncertainty, providing tools for surrogate modeling, Bayesian optimization, and active learning for data-scarce complex systems.

An interactive decision tree visualization tool for Jupyter notebooks that supports scikit-learn, XGBoost, LightGBM, and ONNX models.

A scalable, block-sparse tensor framework for high-performance tensor arithmetic that allows users to compose complex expressions using math-like C++ syntax.

A general framework for benchmarking different vector search engines under the same hardware constraints to compare their performance and efficiency.

A library for measuring and optimizing the energy consumption of deep learning workloads across various hardware platforms including NVIDIA, AMD, and Apple Silicon.

A local-first issue tracker CLI that provides a persistent memory layer and behavioral guardrails for AI-assisted development to prevent context loss across sessions.

A novel neural network architecture that generalizes residual connections using a learnable Delta Operator to dynamically control information erasure and injection across layers.

A Python module for building topological neural networks that enables deep learning on topological domains like simplicial complexes for science and engineering.

A set of Quarkus extensions that integrate the LangChain4j library, allowing Java developers to easily incorporate LLMs, embeddings, and document stores into their applications.

A library that extends the safetensors format to provide a unified storage and management system for various LLM compression and quantization schemes.

LLM‑TPU is an open‑source SOPHGO project that compiles HuggingFace LLM and vision‑language models into SOPHGO TPU‑compatible bmodels. It supports one‑click compilation, a large catalog of quantised models, dynamic/KV‑cache inference, multi‑chip scaling, and provides ready Python/C++ demos for BM1684X, BM1688 and CV186X chips.

A toolkit for evaluating PII detection models and generating synthetic PII datasets using templates to improve Named Entity Recognition (NER) accuracy.

A collection of ML software builds and tools optimized for the deprecated AMD GFX906 architecture to enable modern AI workloads on legacy hardware.