Ph.D. student, Data Science Institute, University of Chicago

I am a third-year Ph.D. student at the UChicago Data Science Institute, working with Haifeng Xu and Nikolaos Ignatiadis. Before that, I did an M.S. in Statistics with Daniel Sanz-Alonso and a B.A. in Statistics & CS at UC Berkeley, with some time as a software engineer in between. I am also part of the AER Lab with Sebastian Musslick.

I work on reinforcement learning for LLM agents and on statistical methods for noisy data.

scholar github linkedin email cv

research

Training agents

Core contributor to rLLM, an open-source framework for agentic RL; making reasoning models stop overthinking (ShorterBetter).

Evaluating models

Building Prophet Arena, which tests LLMs on forecasting real-world events; studying what benchmarks measure and how alignment narrows what models generate.

Prediction-powered inference

Estimators that combine a small amount of reliable data with a large amount of cheap ML predictions, using empirical Bayes ideas for guarantees (PAS).

now

MTS Intern at Stealth Frontier Data Lab, building synthetic environments for agentic RL. Previously a Research Intern at Snowflake AI Research (Mar–Sep 2026), post-training agentic systems.

news

• 2026LLM Probability Concentration accepted to TMLR.
• 2026Empirical Bayes Rebiasing accepted to NeurIPS 2026.
• 2026Bayesian symbolic regression paper accepted to UAI 2026.
• 2026Prophet Arena and benchmark-overlap papers accepted to ICLR 2026.
• 2025ShorterBetter at NeurIPS 2025; PAS at ICML 2025.