Explore the directory

A short visual explanation of how attention works in transformer models, written for students trying to build an intuitive picture of the mechanism.

• Tutorials

A keyboard-first local web application for orchestrating multi-agent software work, letting a developer capture tasks, dispatch coding agents and review their output across several repositories.

• Coding agents

An autonomous terminal coding agent written in Python, with plan, code and chat modes, custom sub-agents, task tracking and evidence-gated memory that persists across long tasks.

• Coding agents

A Model Context Protocol server that lets an AI assistant read, create and modify SAP BW/4HANA or BW 7.5 modeling objects directly, using the same internal APIs as SAP's own Eclipse tooling.

• MCP servers

A framework that follows the AutoML philosophy to let a developer build a customized retrieval-augmented generation pipeline in a few lines of code, without handling each module separately.

• RAG

An eighteen-notebook course that measures each component of the recursive loop behind an AI coding agent, such as verification and state persistence, with reproducible numbers on real benchmarks.

• Tutorials

A structured knowledge repository covering the full lifecycle of building and operating agentic AI systems, from architecture patterns and development frameworks to security, governance and operations.

• Curated lists

A community-maintained hardware plugin that adds an Intel Gaudi accelerator backend to vLLM, so large language models can be served on Gaudi hardware through vLLM's existing serving stack.

• Inference

An in-process agentic memory system for cyber threat intelligence that extracts CVEs, threat actors and indicators from analyst notes into a STIX knowledge graph and serves past cases back through MCP.

• Agent memory
• Security

A family of large language models trained entirely on domestic Chinese computing infrastructure by China Telecom's AI research institute, released with reasoning-focused variants.

• Models

A framework for running neural network models in the browser, React Native and Electron on the user's own GPU, compiling its kernels to WebGPU shaders at runtime instead of a WASM binary.

• Inference
• Libraries

A benchmark that measures whether a local coding agent keeps fixing a bug when repository content, such as a README, an issue or a code comment, tries to redirect it with a hidden instruction.

• AI evals
• Security

A local-first knowledge layer that turns an organization's scattered documents, operational data and decision history into searchable, source-linked context an AI agent can inspect and reuse.

• Agent memory
• RAG

A set of fifteen executable AI-assistant skills that walk a developer through finding, evaluating and submitting their first open-source contribution, across several coding assistants.

• Agent skills
• Tutorials

A 70 megabyte Docker image that runs Ollama on CPU only, built to be far smaller and faster to deploy than the official four-gigabyte Ollama image on systems without a GPU.

• Inference
• Containers

An observability software development kit for multi-agent AI systems, extending OpenTelemetry's generative-AI semantic conventions and exporting to any OpenTelemetry-compatible backend.

• AI evals
• Monitoring

A Fastify plugin that implements a Model Context Protocol server over JSON-RPC, letting a Fastify application expose its own tools, resources and prompts, with an in-process client for testing.

• MCP servers

An open-source authorization engine for AI agents that checks every proposed action against policy before it runs, adding prompt-injection detection and confidence-based human review.

• AI evals
• Security

A zero-dependency C runtime for running massive open-weight Mixture-of-Experts language models on consumer hardware by streaming expert weights across SSD, RAM and VRAM as one memory hierarchy.

• Inference

An AI agent orchestration tool for open-source project maintenance, using a deterministic pipeline to classify GitHub issues and pull requests before dispatching only judgment calls to AI agents.

• Coding agents
• Git

A Claude Code, Copilot and Cursor skill that generates a formatted README for any project through a configure, scan, generate and beautify pipeline, in a choice of writing tones and languages.

• Agent skills
• Documentation tools

A collection of native TouchDesigner operators that expose Apple's on-device machine learning and media frameworks, such as Vision, Core ML and RealityKit, directly inside TouchDesigner.

• Generative media
• Design tools

A tool that reconstructs a path from custom C code to a runnable kernel on Intel Core Ultra NPUs, exposing the chip's programmable SHAVE cores that Intel's public software stack otherwise hides behind graph-level programming only.

• Libraries

A local Russian speech-to-text engine written in Rust, built on the GigaAM v3 model and ONNX Runtime, usable as a command-line transcriber, an HTTP and WebSocket server, or an embedded library from several languages.

• Speech