01AI REVIEW SYSTEM
Research Deliberation
An AI pre-review system that helps researchers expose weak assumptions before they become research designs.
View Project
AI for you
林徽东
I build AI to help peoplethink better,work together,and be understood.
I have worked across public, commercial, and nonprofit settings, learning how different institutions define problems, make decisions, and act under uncertainty. What drew me to AI was not the technology itself, but the possibility of building tools that sharpen judgment, deepen collaboration, and remember what matters.
Selected Work
02SHARED WORKSPACE
SharedGround
A shared task environment where people and AI agents can observe, decide, and act without waiting for turns.
View Project
03PERSONAL AI SYSTEM
Afu OS
A personal operating system for AI — shared memory, operating rules, and project context that help different agents understand the same person over time.
View Project
Latest Notes
You're not using AI. You're organizing it.
The default metaphor for AI is a tool — you pick it up, use it, put it down. But that breaks when you're coordinating multiple agents across days, each with different context, and the real work is keeping state coherent between them.
Why multiple AI roles still produce one opinion
Naming four different reviewer roles in a prompt doesn't create four independent perspectives. They paraphrase each other under different names. What actually changed the output was context isolation — and that took three versions to figure out.
What a failed RAG experiment taught me about AI memory
I built a full RAG pipeline for my personal AI system. The engineering ran perfectly. The answers got worse. The problem wasn't the retrieval quality — it was what "relevance" meant in the first place.