Explore the directory

A free, open curriculum of notebooks, papers and hands-on apps that takes learners from beginner to production AI engineer over 17 weekly sessions.

• Tutorials

A CLI that downloads curated collections of AI agent personas, instructions and prompts, mainly to configure GitHub Copilot for a specific tech stack.

• Agent skills

An object-storage-native key-value cache for LLM inference that lets any engine pointed at the same S3 bucket reuse prefilled context across restarts and machines.

• Inference
• Databases

An LLM inference engine for AMD RDNA4 GPUs written in Rust using Vulkan compute shaders, supporting GGUF models and native FP8 matrix operations.

• Inference

An Android library for streaming, on-device speech recognition, built on a custom Rust inference runtime and a 32M-parameter NeMo FastConformer model.

• Speech
• Mobile

A Mixture-of-Experts language model implemented from scratch in native PyTorch and trained on two free Kaggle T4 GPUs, meant to teach modern LLM internals.

• Models
• Tutorials

A Swift telemetry SDK that traces AI workflows such as inference calls, tool executions and streaming generations, and exports the traces to OpenTelemetry.

• AI evals

A lightweight process supervisor and message router that forwards NDJSON-framed JSON-RPC messages between IDEs, CLIs and worker processes running agent logic.

• Agent frameworks
• MCP servers

A local inference daemon for running LLMs on Apple Silicon that exposes an OpenAI- and Ollama-compatible API, built for fast model loading and low latency.

• Inference

A Squeak/Smalltalk package that adds an OpenAI API client, conversational agents, a ChatGPT GUI, and a built-in vector database for semantic search and RAG.

• RAG
• Libraries

A framework that routes AI agent-to-agent messages through a verifier proxy which authenticates, encrypts and logs every message for later audit.

• Agent frameworks
• Security

A local-first production evidence gate that scans a repository for security, correctness, scale and performance issues without executing its code.

• Code quality
• Security

An open-source framework for running autonomous AI agent swarms on AWS Lambda, scaling to zero cost when idle and fanning out to many agents on demand.

• Agent frameworks

A Rust research prototype for a DNA-inspired, self-evolving control layer around LLM inference, handling routing, memory, reflection, and device adaptation around an existing inference backend rather than being a production inference engine itself.

• Inference

A self-hosted code context engine that indexes Git repositories with tree-sitter and exposes them as MCP tools so AI coding assistants in any editor can search versioned, symbol-aware code context.

• MCP servers
• Agent memory

A RAG parameter-sweep and evaluation toolkit that systematically tests combinations of embedding models, chunking strategies, and retrieval methods against a user's own data and questions, before they build a RAG application.

• RAG
• AI evals

The source for the RadCrew.org website, a monorepo pairing a Vite React frontend with a FastAPI backend that answers visitor questions through a guardrailed, RAG-grounded chatbot.

• RAG

An open-source, Python-first, API-driven Retrieval-Augmented Generation platform with FastAPI endpoints, LangGraph-native agents, a pluggable vector store, and support for several LLM providers.

• RAG

An unofficial command-line client for the Perplexity API, supporting interactive chat, one-off web-search queries, shell completion, and MCP integration for use from other AI tools.

• Inference
• CLI tools

A local-first, Rust-built personal AI assistant workspace with a native desktop app and a self-hosted gateway that owns state, model access, memory, tasks, MCP servers, and skills, runnable on your own machine or infrastructure.

• Agent frameworks

A local, CPU-only agentic AI runtime for Linux that combines chat, model-selected tool use, file workflows, and static artifact analysis in one terminal client, built on a vendored llama.cpp backend.

• Coding agents
• Inference

A gateway-first local model runtime and CLI that pulls, runs, and serves GGUF or Ollama models behind an OpenAI-compatible endpoint, with a dashboard, token telemetry, and an optional GPU capacity marketplace.

• Inference
• LLM gateways

An MCP server that helps an AI chat client find a good open-source "first issue" matched to a developer's skills, using live GitHub data to check whether an issue is already claimed and plan a pull request.

• MCP servers
• Git

A demonstration project showing how to build hybrid search and retrieval-augmented generation with OpenAI embeddings stored in a Pinecone vector database and orchestrated with LangChain, including a PDF-loading pipeline.

• RAG
• Tutorials