Explore the directory

A beginner-friendly guide and automated setup script for running the 350-million-parameter IBM Granite 4.0 language model fully offline on a Raspberry Pi 5 through Ollama, using a GGUF quantization tuned for the Pi 5's CPU.

• Inference
• Tutorials

A local LLM inference proxy built on llama.cpp and llama-swap that lets you load and hot-swap multiple GGUF models and serve them through an OpenAI-compatible HTTP API, with an outbound-connection design for secure remote access to home or office GPUs still on the roadmap.

• Inference
• Networking

A strictly-typed, Go-native alternative to CrewAI for orchestrating parallel AI agents, with LangGraph-style durable flow persistence, agent-to-agent and MCP protocol support, human-in-the-loop approvals, and security-hardened tool execution.

• Agent frameworks

A multimodal semantic search pipeline that ingests documents and images through Apache Airflow DAGs, extracts text and generates image descriptions, embeds the content, and serves cosine-similarity search over a Qdrant vector index with PostgreSQL for metadata.

• RAG
• Data engineering

Reference implementations from Vector Institute's Agentic AI Evaluation Bootcamp, covering agent evaluation fundamentals and four example agents evaluated with LLM-as-judge metrics, batch evaluation and a misalignment probing experiment across several commercial models.

• AI evals
• Tutorials

A set of optimizations for Unsloth on NVIDIA DGX Spark hardware that speeds up LoRA and full fine-tuning of Qwen3.5 models by roughly 7 to 8 times, with a loss curve verified bit-identical to the unoptimized baseline within BF16 precision.

• Models

A documentation set for building and operating a two-node NVIDIA DGX Spark cluster connected by 200 Gb/s direct-attach networking, covering distributed training with NCCL, CUDA inference through EXO, NVMe-TCP shared storage, monitoring and scaling.

• Hosting
• Tutorials

A deep learning framework that classifies 16S rRNA gene sequences into all seven taxonomic ranks in a single pass, combining a convolutional network with a BERT-style transformer and reaching 92.95% species-level accuracy on a public benchmark.

• Models

An idiomatic Go port of the CrewAI framework, with zero external dependencies beyond the standard library, for assembling teams of role-based agents that collaborate sequentially or hierarchically on tasks using any LLM provider.

• Agent frameworks

A self-hosted platform that connects coding-agent harnesses such as Claude Code, Codex or its own Hermes Agent to messaging channels including Discord, Telegram, WhatsApp, Slack and Email, installed and run through a local browser console with offline speech models.

• Agent frameworks

An AI contract review system for enterprise legal and compliance teams that combines retrieval-augmented generation, a multi-agent router and Taiwanese statute lookup to flag risky contract clauses with cited legal sources and suggested edits.

• RAG

A local-first memory tool that compiles Codex and Claude Code session history into an auditable, plain-Markdown wiki so coding agents can recall past investigations, decisions and failures instead of repeating them, without a vector database.

• Agent memory

A multi-LLM AI coding agent, available as a desktop app and terminal CLI, that turns natural-language requests into file edits, code search, shell commands and git operations, decomposing complex tasks across up to 20 parallel sub-agents.

• Coding agents

A one-line-install, container-based sandbox for running coding agents such as Claude Code, Codex, Gemini CLI and OpenCode in isolation, with shared authentication, a controllable built-in browser, and a menu for launching each tool in its auto-approve mode.

• Coding agents
• Containers

A Capacitor plugin that embeds a native Rust AI agent engine directly on Android and iOS, running Claude with file tools, code execution, git and extensible MCP device tools, on-device vector memory and background scheduling, with no cloud relay.

• Agent frameworks
• Mobile

A curated reading list of research on models and agents that improve reasoning and planning by iteratively refining an internal latent state rather than producing explicit intermediate text, spanning supervised reasoners and reinforcement-learned planning.

• Curated lists
• Papers

A curated, pull-request-friendly directory of AI agents maintained as JSON entries, each with its description, supported protocols such as A2A, and links to its homepage, agent card and container image, validated and rendered into this README automatically.

• Curated lists

A GitHub Actions template for a fully self-contained, cron-triggered autonomous agent that uses Claude Code to research, code and open pull requests toward a given goal every day, with PR review, auto-merge and configurable safety limits, and no external servers.

• Coding agents
• Workflow automation

A self-hosted AI platform with a small core (RAG plus knowledge-graph memory, a provider-agnostic LLM gateway with fallback, and governance), a lifecycle manager that deploys, operates and scales the underlying infrastructure, and installable modules built on top.

• RAG
• LLM gateways

A command-line tool that transcribes and summarizes local audio or video files or YouTube content entirely on Apple Silicon using MLX, picking Voxtral or Whisper for transcription and a Qwen3 model for summarization based on the Mac's available memory.

• Inference
• Speech

Research code for a system that answers open-ended questions about artworks by having a multimodal LLM read a painting and a predicted artwork card, then retrieving supporting passages from a multimodal Wikipedia knowledge base when it needs external knowledge.

• RAG
• Papers

A runtime governance proxy that sits between an AI agent and its model API, detecting and blocking prompt injection and instruction-authority transfer from web pages, emails, documents and tool output before the agent acts on it.

• AI evals
• Security

Runnable lesson code from a live AI agents bootcamp taught in Urdu and English, covering LLM API basics, retrieval-augmented generation, building an agent harness from scratch, multimodal agents, LangChain, agent-to-agent protocol, and training a small language model.

• Tutorials

A small Python crash-course repository with runnable examples that progress from a plain OpenAI chat loop to tool-using AgentSpan agents with memory, web search, guardrails, structured output and human-approval steps.

• Tutorials