I am a statistician joining Williams College as an Assistant Professor in July 2024. I received my PhD from the University of California Los Angeles in the Statistics department. During my PhD I developed Statistical methods focused on social science applications with Professor Mark S. Handcock. I previousl worked tackling clinician burnout at Atalan Tech, where I designed, built and maintained a scalable end to end Machine Learning system. I have worked as a research Data Scientist at Meta where I spent my time advocating for, and implementing data driven solutions for privacy and infrastructure related problems.
In particular I am interested in network generation processes, Bayesian social network models, causal inference for social networks and spatial point process models. Please take a look at my research page for publications and working papers. In general I am interested in problems where applying an "off the shelf" method does not yield sensible results. I consider my secret sauce to be the ability to frame a complex problem in a principled manner and drive towards the final outcome of an analysis.
I am originally from the UK and after finishing a masters in Mathematics I trained to be a pensions actuary for two years, which was great grounding in basic statistics and data analysis. I've now moved over the pond after finding that I prefered tackling interesting data driven problems to calculating people's pensions. I have the privledge to spend my time understanding and explaining data to people and hope to continue my career doing just that.
This website gives a couple of examples of small "for fun" projects, as well as a brief description of my current academic research, my CV is available also.
I spend most my free time riding bicycles, sometimes I chase other people riding bicycles in circles too. I like baking cakes and bad sourdough, during the pandemic I slowly transformed my fridge in into a homemade cheese cave. Feel free to contact me with any of the below methods, emails relating to cake, bicycles, and/or cheese may receive quicker replies.