Part of my research is in astrophysical data analysis and I am often applying statistical methods, including likelihoods, Bayesian analysis, and MCMC to physical models and data sets. For example I am searching for ways to reduce the computational complexity of a difficult analysis (see for example this paper), or for methods to combine different data sources to test a hypothesis (this paper).
A long term interest of mine is machine learning and artificial intelligence (I also minored in robotics). Recently I have been working on a project that applies deep learning to an online image analysis platform for biological images. One of my contributions is a CNN based pixel classifier implemented with tensorflow that now runs in production on www.wolution.com. A very early version of my code, using theano, is on github.
Another recent machine learning project of mine was a toy implementation of deep reinforcement learning based on the now famous paper by google deepmind, with a small 20*20 pixel training environment. My code is also on github. It is fun to play with these algorithms.