Explore the directory

A pinned Docker Compose build that serves a 27-billion-parameter Qwen model, quantized to 4-bit AWQ, on AMD's Strix Halo integrated GPU through vLLM with speculative decoding and tool calling.

• Inference
• Containers

A post-retrieval layer for RAG pipelines that filters out expired documents, applies time decay and freshness scoring, and reranks results so a fresh, still-valid document outranks an outdated one.

• RAG

A command-line tool for running teams of AI agents, called squads, that share a goal and a persistent memory stored entirely as markdown files in a git repository.

• Agent frameworks
• CLI tools

A personal collection of benchmark results for small language models and quantizations that fit on a single NVIDIA DGX Spark machine, comparing coding and evaluation scores across several models.

• AI evals

A dependency-free command-line tool that runs the same prompt against locally installed Ollama models, repeats it a chosen number of times, and saves every response to disk for comparison.

• AI evals
• CLI tools

An open source security layer for AI agents that enforces role-based access control, requires human approval for high-risk actions, and writes a cryptographically signed, tamper-evident audit log.

• AI evals
• Security

A step-by-step guide to building a personal knowledge base out of plain markdown files, using Claude Code to ingest clipped articles into an Obsidian vault that links new material to what exists.

• Tutorials

An open source Rust engine for building AI companion characters that keep a persistent memory of the user, an evolving relationship model, and a decision layer that keeps a persona in character.

• Agent frameworks
• Libraries

A tool for scoring local and cloud language models against custom prompts using a configurable judge model, with a bundled web dashboard for building prompts and browsing a leaderboard.

• AI evals

A unified Go command-line tool that manages configuration, instructions, skills, plugins, MCP servers and launch settings for seventeen AI coding assistants from one terminal interface.

• Coding agents
• CLI tools

An OpenAI-compatible chat server that puts any Hugging Face model behind a normal chat API while also showing a live view into what the model is doing internally as it generates a reply.

• Inference
• Debugging

A searchable catalog of more than 300 runnable "/goal" contracts that give a coding agent a narrow, verifiable task with explicit constraints, a definition of done, and stop rules for uncertainty.

• Agent skills
• Curated lists

A complete system-on-chip project implementing a quantized convolutional neural network in custom hardware on a twenty-five-dollar FPGA board, classifying blood-cell images faster than an interpreted Python loop on a laptop CPU.

• Inference

A companion set of computational notebooks, exercises and projects for the "No Bullshit Guide to Statistics" book, meant to be run in JupyterLab rather than only read.

• Tutorials
• Books

A short set of tutorials on using machine-learning methods for causal inference.

• Tutorials

A Unity Editor tool that lets an AI assistant inspect and modify Unity's ParticleSystem component through natural language, covering emission, shape, velocity, color and all 24 particle modules.

• MCP servers

A fork of the turbovec vector search library that adds a graph-memory layer, letting a single query filter retrieval results by tenant, tag, source and time alongside vector similarity.

• Agent memory
• Databases

A two-board embedded AI demo that runs a 28.9-million-parameter language model directly on an ESP32-S3, with a separate RP2040 acting as a serial OLED display node for the generated text.

• Inference

A Docker Compose setup that runs the pi coding agent in an isolated container with file-system isolation, dropped privileges and persistent storage, including an offline mode using a local llama.cpp model.

• Coding agents
• Containers

A four-layer, self-paced roadmap that takes a Java developer from programming fundamentals through JVM and Spring internals to an AI-engineering track built on Spring AI, RAG, agents and MCP.

• Tutorials

A starter git repository for creating a persistent, autonomous AI agent built on gptme, giving the agent its own workspace, identity file and architecture notes that it can read and help extend over time.

• Agent frameworks

A wire format for structured data exchanged between LLMs and tools, designed to need far fewer tokens than JSON while staying schema-free, lossless, and readable by a model without special training.

• Agent frameworks
• Libraries

An operating environment for long-running AI agent work that keeps plans, progress and decisions in files and Git, so a task can hand off across sessions and tool changes while a person reviews at set points.

• Coding agents
• Agent frameworks

A Claude Code skill that interviews the developer and scaffolds a ready-to-run LangChain Deep Agents agent, or a dcode terminal coding agent, into the current project.

• Coding agents
• Agent skills