I have used analytic calculations and analyzed the output of large numerical simulations of structure formation to sharpen the theoretical interpretation of data from large galaxy redshift surveys to answer fundamental physics questions.   In the last several years I have also been deeply involved in the analysis of real survey data -- dealing with complex geometrical masks, proper error estimation, observational constraints that remove a non-representative subsample of objects (“fiber collisons”), and observational systematics (like looking through the stars and dust in our own galaxy) that could be confused with the cosmological signal we’re after.

In my day-to-day work as a cosmologist I’ve gained experience in several programming languages, math libraries, and data analysis techniques and packages.  Details are listed below.  Some examples are available in my Code/Data section.

While I use python daily for data analysis and plotting, most of my serious coding is done in C.  I make extensive use of the GNU Scientific Library for problems like multidimensional nonlinear least squares fitting, multidimensional integration, root finding, high quality random number generation, matrix operations, and evaluating special functions.  During my work with the Atacama Cosmology telescope collaboration, I wrote the code to solve for the pointing matrix and point spread functions of ~1000 detectors using time series data of point source observations.  Our C calculations were interfaced to python using swig.

I have become proficient enough in Mathematica, Fortran 90, IDL, and SQL to complete a few research projects.

Markov Chain Monte Carlo is a popular method for exploring large parameter spaces; I have contributed to the popular cosmology package cosmomc as well as written my own MCMC sampler.