Explore the directory

A local research workflow for capturing a language model's internal activations and directly editing its weights to suppress a chosen behavior, without any training step.

• Models

A self-hostable AI memory layer, API-compatible with Mem0, that stores conversations as searchable long-term memory using a FastAPI service in front of a Qdrant vector store.

• Agent memory

A native macOS app that wraps the Pi coding agent's CLI sessions in a SwiftUI workspace, giving each session and subagent its own visible, isolated git worktree.

• Coding agents

A Rust framework for building and managing autonomous AI agents, with declarative YAML agent configuration, long-term memory, planning and multi-agent collaboration.

• Agent frameworks

An LLM-powered Slack bot that answers questions using a company's own Slack history, GitHub pull requests, and Linear issues, run on a Postgres and TimescaleDB backend.

• RAG
• Databases

A distributed multi-agent dialogue simulator, built on FastAPI and asyncio, where language model agents negotiate meaning and build a shared symbolic vocabulary through recursive interaction.

• Agent frameworks

A zero-trust sandbox framework that limits which files and commit history an AI coding model can see per session, logging every prompt, response, and data transfer.

• Security
• Containers

An MCP server that gives AI coding agents a shared, persistent memory backed by Postgres, so context carries over between sessions and between different agent clients.

• MCP servers
• Agent memory

A skill that adds a stronger, specialized reasoning model as an architecture and security advisor for AI coding agents, examining a codebase's structure before code is generated.

• Agent skills
• Code quality

A fork of llama.cpp tuned specifically for NVIDIA Ampere consumer GPUs such as the RTX 3090, adding a custom KV cache format and speculative decoding for large context windows.

• Inference

A local knowledge engine that indexes a folder of notes, code, PDFs, and Office documents, then answers questions with citations through a CLI, web UI, REST API, or MCP.

• RAG

A worker service that runs polls as CSV files in a GitHub repository, using commits as ballots and GitHub OAuth to validate who is allowed to vote.

• Git
• Backend

A Swift SDK that downloads, caches, and runs Apple's Core AI models for chat, vision, and speech in an iPhone or Mac app, without a Python conversion step or bundled weights.

• Inference
• Mobile

A command-line tool that sits between code generation and formatting for Swift, applying rules such as adding access modifiers or memberwise initializers based on the parsed syntax tree.

• Code quality

A plugin for the OpenClaw AI assistant that classifies each incoming request and routes it to a 4-bit, 8-bit or 16-bit precision model target, trading quality for latency and cost automatically instead of asking the user to choose.

• LLM gateways

A text language for describing human movement that can be parsed, validated and rendered in the browser as 3D animation, keeping the semantic decisions behind a motion readable instead of hidden in a black-box trajectory.

• MCP servers
• Languages

A native desktop chat application, built with Rust, Tauri, and Candle, for running large language models locally with no cloud connection or account.

• Inference

An open-source orchestration runtime that treats AI agents, tools, and policies as declarative YAML resources, scheduling, routing, and governing them like production infrastructure.

• Agent frameworks

A source-available, self-hosted runtime for running fleets of autonomous AI agents in isolated containers, with a central credential vault and per-agent spending budgets.

• Agent frameworks
• Containers

A command-line AI coding agent built for small, free, or local language models with 8000-token context windows instead of large paid models.

• Coding agents

A cookbook repository of integration patterns for combining several AI APIs, such as GPT, Claude, Gemini, Sora, Suno, DeepSeek, and Kimi, into one workflow rather than calling a single model.

• Tutorials

A research system, presented at OSDI 2025, for running large language model inference on wafer-scale AI accelerators instead of GPU clusters.

• Inference
• Papers

A from-scratch large language model stack, covering the model, an instruction set, a compiler, a simulator, and a GPU design, built small enough for one person to read end to end.

• Models
• Tutorials

An open-source backend runtime, AI orchestration layer, and command-style web interface for building simulation environments such as supply-chain or crisis-response training exercises.

• Agent frameworks