Tencent Hunyuan LLM Frontier, Seattle — Research Scientist
Ph.D. in Computer Science, University of Georgia
Advised by Prof. Ninghao Liu
I work on agent reinforcement learning and self-evolving AI —
building agents that improve through environmental interaction and self-generated experience.
My current focus is agents for long-horizon terminal work: acting through real tools, training
on executable environments, and learning from the trajectories they create.
In summer 2026 I am leading a research program on self-evolving terminal agents,
working with Zhongzhi Li, Ruhan Wang, Zongxia Li, and Junyao Yang. We study how an agent training
system can improve along three coupled axes: the environments it trains on, the harness through
which it acts, and the learning signals extracted from long trajectories. Notes and artifacts are
released as the work matures.
Forty-six long-horizon tasks across nine categories, with dense intermediate rewards that measure meaningful progress through sustained terminal workflows.
The best way to reach me is email: mailofsyc@gmail.com.
I am always glad to talk with people working on self-evolving agents, long-horizon interaction,
and agent post-training.
