Software engineer · AI agent tooling · Quantum measurement & control
I'm a software engineer based in China, building dependable tools for AI coding agents: project-scoped skills, model routing, terminal observability, and ways to bring work into everyday workflows.
My background is in quantum measurement and control, from chip topology to calibration workflows. Across both domains, I care about explicit boundaries, observable behavior, and software that engineers can inspect, test, and use.
Recent merged upstream work includes workspace-scoped skill discovery in cc-connect, Kimi conversation history in herdr-remote, and an MCP SDK security update in pi-web-access.
- 🧪 OpenAI Agents SDK Learning Lab — a hands-on path from a first agent to a bounded, observable terminal application.
- 🌱 Learn QEC — learning quantum error correction, surface codes, and lattice surgery in public.
- 🧭 QArray — exploring geometry-first representations of quantum chip topology.
- Start with a reproducible problem, clear boundaries, and explicit failure behavior.
- Keep changes reviewable; check them against real agent hosts, provider interfaces, and recovery scenarios.
- Build tests, observability, and documentation alongside the tool.
Ask me about quantum chip topology, QEC learning, dependable agent systems, or developer tooling. If something I build is useful—or could be better—the simplest way to reach me is through the relevant repository or here as @M1racleShih.