A minimalist, open-source LaTeX OCR model with only 20M parameters that converts mathematical formula images into LaTeX code.

An interactive visualization tool for exploring unstructured datasets, helping ML teams identify data issues and critical clusters across multiple modalities.

A collection of easy-to-read, standalone PyTorch implementations of core reinforcement learning algorithms, from basic Grid World to deep RL for Atari games.

A multi-provider gateway that allows coding agents to use various LLM providers and local models via a unified proxy with automatic fallback routing.

An open-source mobile app for finding, reading, and summarizing arXiv papers using AI-powered chat and summaries.

ScreenCoder is a multi-agent system that transforms UI screenshots and design mockups into production-ready HTML/CSS code using visual understanding and layout planning.

An LLM server and client optimized for Rockchip RK3588/3576 platforms that enables AI models to run on the NPU for accelerated local inference.

An enterprise-grade AI marketing automation toolkit providing specialized agents, commands, and skills for AI assistants like Claude Code and Cursor.

A semi-automatic graphic annotation tool for OCR that uses built-in PP-OCR models to automatically detect and recognize text for training dataset creation.

An open-source API security platform that provides continuous API inventory and tests for business logic vulnerabilities using real traffic data.

A JavaScript acceleration library that transpiles JS functions into shader language to enable GPGPU (General Purpose computing on GPUs) in the browser and Node.js.

A Python library that provides a unified interface to extract text from various document formats, simplifying the data preparation process for text analysis.

An open-source project providing full hardware and firmware designs for high-performance brushless servo actuators used in robotics.

A library that brings WebGPU capabilities to React Native applications using the Dawn implementation for high-performance GPU computing and graphics.

A secure JavaScript runtime designed for edge computing and AI workloads that uses WebAssembly for sandboxing to ensure safe execution.

A scalable Python toolkit for analyzing single-cell gene expression data, supporting preprocessing, clustering, and trajectory inference for datasets with over a million cells.

A monitoring and control package for NVIDIA Jetson series boards that provides real-time hardware stats and system management via a terminal UI or Python library.

A macOS utility that uses OCR to extract non-selectable text from any screen area and copy it directly to the clipboard.

An automated AI news aggregation and analysis platform that uses Cloudflare Workers and Google Gemini to generate and publish daily AI reports to GitHub Pages.

A macOS CLI tool for on-device speech transcription of files, live system audio, and microphone input using Apple's Speech.framework.

A Python library for programmatically downloading, uploading, and loading Kaggle datasets, models, and competition resources into data science frameworks.

A community-driven standard providing a catalogue of testable security requirements to verify and secure AI-enabled systems across their entire lifecycle.

A primitive parallel-formant speech synthesizer that recreates the retro sound of 1970s and 80s computer speech for the browser, Node.js, and CLI.

A collection of custom plugins and skills for Claude Code that adds specialized capabilities for product management, agentic system design, and automated technical workflows.

Conversation Stenography is an open‑source Go CLI that encrypts a secret message with AES‑SIV, encodes the ciphertext into token choices of a locally‑run LLM (e.g., Llama 3.2 or GPT‑2), and outputs natural‑looking cover text. Users copy that text into any chat app; the recipient runs the same tool to decode and decrypt the hidden message. The system is fully offline, uses a shared secret phrase (PBKDF2‑derived key), and chains messages for integrity. It includes a setup wizard, simulation mode, and several CLI commands, but is a proof‑of‑concept and vulnerable to existing steganography detection methods.

A mobile and desktop interface for AI coding agents like Codex and Claude, allowing users to to manage projects and git workflows from their own local machines via a bridge server.

M3DGR is a multi-sensor SLAM benchmark dataset and resilient framework designed to test and improve the robustness of ground robot localization and mapping in degraded environments.

A template for a team of coordinated AI marketing agents that automate brand positioning, content creation, and distribution across Notion, Typefully, and Resend.

A collection of open-source Three.js experiences demonstrating interactive 3D showrooms, games, and data visualizations built with the Mint asset pipeline.

A modified version of TOKENICODE that provides a desktop GUI for Claude Code and DeepSeek, featuring enhanced provider mapping, a Claude Skills management panel, and integrated web previews.

A multisensor-aided inertial navigation system that flexibly fuses IMU, wheel encoders, cameras, GNSS, and LiDAR for robust robot localization and online sensor calibration.

An LV2 plugin for playing back neural network machine learning amplifier models using the NeuralAudio engine.

A durable, versioned workspace provider for disposable AI agent sandboxes that ensures artifact lineage and state survival through atomic metadata control.

A specialized knowledge base for AI coding agents that provides expert guidance on modern Swift Testing APIs, XCTest migration, and parallel test execution.

A Python package for estimating heterogeneous treatment effects from observational data by combining machine learning with econometrics for automated causal inference.

A collection of example programs and libraries for MLX Swift, demonstrating how to build LLM, VLM, and image generation apps on iOS and macOS.

A minimal, dependency-free C library for training and using feedforward artificial neural networks with support for backpropagation and alternative optimization methods.

An opinionated agent infrastructure that uses a multi-process architecture to handle concurrent conversations, long-term typed memory, and autonomous background tasks for teams and communities.

A command-line tool for agentic document extraction that converts complex documents into grounded Markdown and structured data with verifiable evidence.

A collection of lightweight, discrete grid-world environments for Reinforcement Learning research, supporting goal-oriented tasks and grounded language learning.

An AI workstation that enables autonomous AI agents to operate a local browser, terminal, and knowledge base to automate complex workflows 24/7.

A local-first scientific writing workspace that integrates LaTeX, Python, and Claude AI to help researchers write papers and perform data analysis offline.

A skill-centric agent runtime that allows AI agents to design, execute, and persist reusable skills to improve their performance on complex tasks through a reflection-based learning loop.

A web-based visual interface and logger for Claude Code that adds a GUI, code diffing, and mobile access to the terminal-based AI coding tool.

A high-performance operator library implemented in Triton that enables LLM training and inference to run across diverse AI accelerators using PyTorch APIs.

A highly extensible AI bot framework that enables LLMs to execute code in secure sandboxes and interact across multiple chat platforms via a unified adapter architecture.

A suite of test scenarios and games for multi-agent reinforcement learning to evaluate how AI agents generalize social interactions to novel situations.

A Microsoft Azure‑based reference solution that demonstrates how to build a multi‑agent AI orchestration platform for automating complex business workflows. It combines Azure OpenAI (Foundry), Container Apps, Cosmos DB, and other Azure services, and includes deployment guides, cost guidance, and security recommendations.