Bio

I am a second-year PhD student in Statistics at UC Berkeley, supported by an NSF Graduate Research Fellowship. My work brings tools from causal inference and machine learning to the study of large AI models, both to understand their capabilities from first principles and to turn them into trustworthy tools for statistical inference and decision-making. Recent work studies which learning algorithms transformers can implement through continuous latent reasoning, and how structured attention masks trade off expressivity and efficiency in long-context models.

Before Berkeley, I received a BS in Applied and Computational Mathematics from Caltech in 2023. I then spent two years as a quantitative trader at the Voleon Group, a machine-learning hedge fund, where I worked on portfolio risk, automated hedging, and securities lending. These experiences drew me to the foundational questions behind the statistical tools I used every day.

Education

• PhD in Statistics, University of California, Berkeley, 2025–present
• BS in Applied and Computational Mathematics, California Institute of Technology, 2023