Hello!

I am a postdoctoral researcher at Harvard University’s Computer Science department and the Harvard Data Science Initiative, co-supervised by Milind Tambe and Francesca Dominici. My current focus is on training agents with foundation models using methods from reinforcement learning (RL), diffusion generative models, and large language models (LLMs). On the applied side, I am interested in autonomous agent systems and social impact applications.

Research

My current projects focus on three directions: first, leveraging LLMs for reinforcement learning, particularly for improving explainability, generalization, and as engines for generative agent-based simulations; second, leveraging diffusion generative models as powerful engines for causal reasoning and planning; and third, developing new representation learning methods for decision-making tasks. Check my Google Scholar for more details about my previous work.

Bio

I hold a BSc and Masters in Mathematics from ITAM and The University of Cambridge. I completed my PhD in Statistics at UT Austin focusing on reinforcement learning and computer vision applications, where I was also a member of the Learning Agents Research Group at the Computer Science Department. During this time, I participated in the UT Austin Villa Robot Soccer Team, where I developed a deep-learning vision system for autonomous soccer robots and competed at the Robocup. Additionally, I held internships at Meta AI (FAIR) and Intel AI, and served as a core member of UT Austin’s Covid-19 response team during the pandemic.

2022 Robocup competition, Bangkok, Thailand