Starred repositories
Claude Autoresearch Skill — Autonomous goal-directed iteration for Claude Code. Inspired by Karpathy's autoresearch. Modify → Verify → Keep/Discard → Repeat forever.
Claude Code skill for autonomous, goal-directed iteration. /autoresearch <goal> builds a real-data benchmark harness, captures a baseline, and iterates with a regression gate until the goal is hit.…
6
Updated Apr 19, 2026
AI agents running research on single-GPU nanochat training automatically
A curated list of autonomous improvement loops, research agents, and autoresearch-style systems inspired by Karpathy's autoresearch.
Agent-native Terminal Multiplexing for 10,000x hyperengineers
The Swiss Army Knife of Offline AI. Chat, Speak, and Generate Images - Privacy First, Zero Internet. Download an LLM and use it on your mobile device. No data ever leaves your phone. Supports text-…
Coding Agent singularly focused efficiency and context curation. Reduces API costs by 50-80% vs other agent AND improves the code quality at the same time. Uses Hash Anchored edits, massively paral…
🪐 Markdown with superpowers: from ideas to papers, presentations, websites, books, and knowledge bases.
WUPHF is a collaborative office of AI employees who build and maintain their own knowledge base to never lose context for the tasks you give them. Supports Claude Code, Codex, OpenClaw and local LL…
Generates metadata.json to mount a tar file in WORKERFS
SQLite extension + bindings for Postgres NOTIFY/LISTEN semantics with durable queues, streams, pub/sub, and scheduler
🤗 ml-intern: an open-source ML engineer that reads papers, trains models, and ships ML models
Instant, Concurrent, Secure & Lightweight Sandbox for AI Agents.
A modern language that compiles to JavaScript
A SOTA quantization algorithm for high-accuracy low-bit LLM inference, seamlessly optimized for CPU/XPU/CUDA, with multi-datatype support and full compatibility with vLLM, SGLang, and Transformers.
Local context manager for Claude Code and Codex with workstreams, transcript binding, and branching.
Lucebox optimization hub: hand-tuned LLM inference, built for specific consumer hardware.
Write HTML. Render video. Built for agents.
A self-hosting systems language that speaks TLS alone. Zero C dependencies. GC in ARM64 assembly. Pure-Rail TLS 1.3 + X.509 chain validation in v3.0.0. 4 backends. BSL 1.1.
MDV — a Markdown superset for documents, dashboards, and slides with embedded data and visualizations. HTML + PDF export, live preview, VS Code extension.
Autonomous research loops. Reproducible, shareable, verifiable.
Inference throughput optimization helix for causal language models on WikiText-2