You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Full harness system - knowledge base + tools + skills + onboarding design
Project Goal
This project aims to draw upon the automated research paradigm championed by Andrej Karpathy and others—as well as the concepts behind the recently open-sourced "AI Scientist"—to deeply integrate the "Auto-research" capabilities of Large Language Models (LLMs) into the underlying physics of quantum error correction. The objective is to construct a fully autonomous "AI Decoding Engine" (AutoQEC). Taking the complex noise profiles and inference latency constraints of specific hardware as its inputs, the system leverages agent-based workflows to drive a Network Architecture Search (NAS) process, constrained by the Tanner graph structures of quantum codes. Operating within an unattended, closed-loop workflow, the engine automatically orchestrates Stim to rapidly synthesize massive datasets of noisy syndromes, while autonomously executing the entire cycle of "architecture conceptualization, closed-loop training, threshold evaluation, and self-iteration."
Planned Components
Knowledge base, examples, or domain notes
Skill file or agent instructions
Command line scripts, Makefile targets, or a small CLI
MCP server or external service integration
Tests, checks, evaluation cases, or review prompts
Setup guide, tutorial, or onboarding flow
Demo repo, sample data, or reproducible example
Before-Event Plan
configure coding agent, install skills
Target AI Coding Tool
Claude Code
Participation Readiness
I have access to a Bash or Zsh terminal. Windows users can use WSL 2.
I have installed Codex CLI, Claude Code, or another AI coding tool.
I have tried logging in to my AI coding tool before the event.
I have a concrete workflow, sample input, or starter repo to work on.
Primary Harness Type
Full harness system - knowledge base + tools + skills + onboarding design
Project Goal
This project aims to draw upon the automated research paradigm championed by Andrej Karpathy and others—as well as the concepts behind the recently open-sourced "AI Scientist"—to deeply integrate the "Auto-research" capabilities of Large Language Models (LLMs) into the underlying physics of quantum error correction. The objective is to construct a fully autonomous "AI Decoding Engine" (AutoQEC). Taking the complex noise profiles and inference latency constraints of specific hardware as its inputs, the system leverages agent-based workflows to drive a Network Architecture Search (NAS) process, constrained by the Tanner graph structures of quantum codes. Operating within an unattended, closed-loop workflow, the engine automatically orchestrates Stim to rapidly synthesize massive datasets of noisy syndromes, while autonomously executing the entire cycle of "architecture conceptualization, closed-loop training, threshold evaluation, and self-iteration."
Planned Components
Before-Event Plan
configure coding agent, install skills
Target AI Coding Tool
Claude Code
Participation Readiness
Team / Contact
Jiahan Chen, jiahanchen527@gmail.com
Tengxiang Lin, tengxianglin23@gmail.com
Jingu Xie, jinguxie2021@163.com