Winner 🏆 (Agent-only) MLSys 2026 - FlashInfer AI Kernel Generation Contest for the DeepSeek Sparse Attention (DSA) track with an average speedup of 34.93x

A minimal autonomous code-agent CLI built with Qwen Code

AIDE: AI-Driven Exploration in the Space of Code. The machine Learning engineering agent that automates AI R&D.

A community curated collection of AI agent failure modes and battle-tested solutions.

HY-Embodied: Embodied Foundation Models for Real-World Agents

AgentENV (AENV) is a distributed platform for running agent environments at scale.

Composite-Bench: deterministic, optimality-certified computer-use benchmark

cli terminal app for slack with sqlite backend

Create, develop, and deploy Slack apps from the command-line ✨

The new Windows Terminal and the original Windows console host, all in the same place!

The lightweight framework for building agents

Power BI CLI for the REST APIs

FrontierPhysics: Benchmark how AI agents do frontier physics research.

📡 Your own AI-powered news radar. Generates daily briefings in English & Chinese. | 用 AI 构建你专属的新闻雷达

GenesisBench: How language intelligence can be used to improve physical intelligence.

An awesome list of websites about minimal web design and copious swearing

Multiplayer, self-hosted, secure agents.

the runtime your coding agents live on

OpenTag

SWE-chat: Coding Agent Interactions From Real Users in the Wild

Google Workspace in your terminal.

An environment for runtime hooking/code patching within Super Smash Bros Ultimate

The theory of mind module for the SWE agent

A Claude skill for writing cold outreach that actually gets replies — on email, LinkedIn, or X DMs. Short, specific, human, easy to say yes to.

BenchPress: calibrated LLM benchmark score completion

Skills for Real Engineers. Straight from my .agents directory.

DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms