About Me

I’m a 3rd year PhD candidate at Duke University in the Statistical Science department. I work on statistical and machine learning methodology for high-dimensional data, where difficulties arise from complex dependency structures. Examples include sampling/generative modeling (diffusion models for protein design, image generation) and adaptive experimentation (multi-armed bandits on networks, SUTVA violations).

Addressing these problems requires accurate and efficient inference algorithms. Taking a Bayesian approach, my research centers on the design and analysis of advanced sampling algorithms such as sequential Monte Carlo or adaptive MCMC methods. I have significant experience with the implementation of such algorithms through parallelization on high-performance computing clusters.

Before arriving at Duke, I worked as a Senior Research Analyst at the Federal Reserve Bank of New York. There, I contributed to the development of SMC.jl and DSGE.jl, open-source Julia packages for estimating large state-space models using sequential Monte Carlo.

Outside of research, I enjoy backpacking, reading, and running.