I am a data scientist at Gartner on the Digital Client Experience - Search team, where I work on the Learning to Rank (LTR) model behind the search results ranking. My work covers create novel relevancy features for the model based on current research. The newest feature I built was an embeddings-based recommendation model based on clients' document reading history to aide in predicting which documents the user will be more likely to read in their next search. Each change to the model is shown to be an improvement over the previous after rigorous offline evaluation and online A/B testing.
Previously, I was a data science fellow within the Insight Data Science program, where I worked on predicting how subway riders would change their habits following a station closure. I studied nuclear astrophysics at the University of Notre Dame during my doctoral program, where I focused on studying radiative-capture reactions using the St. George recoil mass separator and exploring alternate uses for the machine. I graduated from Michigan State University in 2011 with degrees in physics and astrophysics and a minor in mathematics from Lyman Briggs College.
I code in Python, both for large-scale analysis and short scripts. Much of my research code has been in C++ extended by the ROOT Data Analysis Framework, especially for work that reaches beyond my own dissertation. I have also used bash scripting, Fortran, COSY, and others when necessary. My current body of work is within data analysis and visualization and equipment monitoring.
My current work in closed-source, so below are a selection of projects I created in support of my dissertation, for my lab and collaborations, and extra projects outside of physics.