Direction
Multi-agent systems
Designing capability boundaries, memory, collaboration protocols and verification loops for AI agents.
杨青峰 Qingfeng Yang
歌贼王·@songth1ef
AI application engineer · turning agents into systems that ship
I build AI applications and agent systems. Day to day that means wiring model capabilities into real business systems — frontend, backend, deployment — end to end, on my own.
Core stack TypeScript (Next.js / NestJS / Vue 3). Alongside that I work on multi-machine agent collaboration: memory shared across hosts, async agent-to-agent protocols, and machine-checkable task acceptance — already running on real production tasks.
What I care about is verifiability and automation leverage: whether a system can tell if it got the job right matters more than whether it runs at all.
Current focus
Projects
Testing judgment through real projects
Nestwork, A2A mesh, cheat-on-content and Duovid are experiments from different angles.
Writing
Turning experience into reusable assets
The writing is not a pile of takes. It records trade-offs, failures and methods from real projects.
Skills
Links
Recent writing
All →Slimming Down Agent Memory: Resident vs. On-Demand
As memory piles up, the amount an agent reads at every startup grows too, until it "remembers everything and can use none of it." My approach splits memory into two tiers. The resident tier holds only "who I work for, what we're doing now, and what moved recently," a dozen-or-so KB in total. Everything else is topic files retrieved on demand. On my own store of about 180 memory files and over 300k tokens, startup loading fell from about 70k tokens to a few hundred. This post covers the tiering rules and what must never go into the resident tier.
Months Without Writing a Line of Code: How I Measure My Output
I haven't written a line of code by hand in months. All the code is generated by AI and committed under one git identity. That breaks every old yardstick, since most commits come from machines and line counts mean even less. So I switched to a new one. I look only at deliverables, ask where the output went rather than how hard I worked, and manage "verify, collect, push, merge" as the real bottleneck. This post covers that yardstick and a few things it forced me to admit.
Managing a Dozen Worktrees Without Knowing Git Well
Git is a clear weak spot of mine. Multiple branches, rebase, conflicts, and reading the commit graph are all unfamiliar, and I've caused real accidents. Yet I now run a dozen-plus worktrees with several agents working on different branches, and things are clearer than before. The trick wasn't mastering git first. It was letting AI handle git while I own a few disciplines it must follow and a few questions I can ask at any time.