Zeyi Chen

I am an undergraduate student in Electrical and Computer Engineering at the UM-SJTU Joint Institute, Shanghai Jiao Tong University. I was an exchange student at Carnegie Mellon University in Spring 2026.

My research interests lie in AI agents, multi-agent systems, agentic web infrastructure, reinforcement learning, and social dynamics. I am especially interested in building persistent, adaptive AI systems that can coordinate with humans and other agents over long time horizons.

Research Interests

AI Agents and Agentic Web. I study how LLM-based agents can use tools, communicate, remember, and coordinate in open-ended web environments. My current work is connected to Holos and broader agentic web infrastructure.

Multi-Agent Systems and Agent Societies. I am interested in how individual policies and interaction protocols give rise to collective behavior, including cooperation, competition, and emergent social patterns.

Reinforcement Learning for Emergent Behavior. I have worked on reinforcement learning problems in mixed human-autonomous traffic systems, with a focus on how adaptive agents shape system-level behavior.

Human-AI Interaction and Long-Term Autonomy. I care about AI systems that remain useful, interpretable, and steerable as they operate across longer time horizons and richer social contexts.

Selected Research Projects

• Holos / Agentic Web Infrastructure — web-scale multi-agent infrastructure for persistent LLM-based agents and human-agent collaboration.
• Mixed Human-Autonomous Traffic Systems — reinforcement learning and simulation for emergent driving behavior in traffic systems with both human and autonomous vehicles.
• Opinion Dynamics in Agent Societies — modeling how interaction structures and adaptive agents influence belief formation and social dynamics.

Looking Ahead

I am developing a research agenda around persistent AI agents, multi-agent coordination, and adaptive systems. I am happy to connect with researchers, builders, and students working on agents, reinforcement learning, social simulation, or the future of human-AI collaboration.