I am a Computational Data Science Fellow at the Berkeley Center for Cosmological Physics at UC Berkeley. I develop statistical and numerical methods for analyzing the exponentially growing amount of observational data in modern and future cosmological surveys. My current focus is large scale structure. My other interests include the theory of inflation and observational tests through the cosmic microwave background radiation data. On the data science side, my interests include Bayesian statistics, sampling and optimization methods for very high dimensional parameter spaces, and machine learning.

I received my PhD in Physics at UC San Diego in 2012, after which I moved to the University of Auckland in New Zealand as a postdoctoral researcher in Early Universe Cosmology. I moved back to California in 2015 for my new position at UC Berkeley.

I am an avid C++ programmer and a python enthusiast, with many years of experience in industry and science. I am the author of the powerful numerical library COSMO++.

You can access my CV here.


The structures we see today (stars, galaxies, clusters of galaxies) have formed through highly nonlinear processes of gravitational collapse, as well as complex astrophysical processes. However, at high redshifts (i.e. the early epochs of the evolution) the universe was much smoother, with very small fluctuations in the matter density. It is much easier to theoretically study these earlier epochs, where the evolution can be very well described with linear theory, compared to the current epoch. In my research I develop methods for accurately reconstructing the initial linear density field from current observations, such as galaxy surveys, weak lensing, and the Lyman-alpha forest.



My current position does not involve any teaching. Please check the links below for previously taught courses.

Student reviews at UC San Diego can be accessed here.

See also: my introductory notes on probability theory for students with no probability/statistics background.


Coming soon.