Muxin (Murphy) Tian

Hi, I’m Murphy Tian, an undergraduate student at University of Toronto. I’m currently a research intern at Microsoft Research Asia. I am fortunate to work with talented research advisors: Prof. Jiaxuan You at U Lab on agentic reinforcement learning and Prof. JunCheng Yang at HarvardSys Lab on LLM serving systems. I have open-source experience as a member of OpenManus and a Google Summer of Code contributor to ChromiumOS, along with industry engineering background from working at Microsoft and RedNote. I was previously advised by Dr. Daniela Rosu on interpretable agentic systems.

Email: murphy[dot]tian[at]mail[dot]toronto[dot]ca

My research focuses on building more capable intelligence through two complementary directions: (1) Post-training and agentic systems — exploring reinforcement learning methods for agent capabilities, frameworks for automatic agentic system generation and optimization, and interpretability and explainability for agentic systems; (2) Efficient and reliable machine learning systems — building infrastructure for inference optimization, scheduling, and resource utilization to make large language model systems more accessible and sustainable.

Feel free to reach out to explore potential research opportunities.