Summer Associate -
In the summer of 2019, I was a summer associate at the RAND Corporation in the Pittsburgh office. I worked in the Statistics Group where I applied machine learning and clustering methods to health-care data. Specifically, I worked with Medicare and Medicaid claims data to determine the similarity of practices.
Voting Rights Data Institute -
In the summer of 2018, I had the great opportunity to work at the Metric Geometry and Gerrymandering Group (MGGG), which is a partnership between Harvard, MIT, and Tufts. I created innovative models on voting behavior. I also worked on compactness and combinatorial models to test for the presence of gerrymandering in district maps.
Computational Social Science
Later in the summer of 2018, I traveled to Bremen, Germany where I participated in the Computational Social Science Research Incubator/Summer School hosted by the Bremen International Graduate School of Social Sciences (BIGSSS). Through this opportunity, I took classes and workshops with my colleagues on topics such as spatial statistics, network analysis, agent-based modeling, and text analysis.
Throughout the duration of the research incubator, we also worked on a research project where investigated new techniques to model civil conflict over time and space.
Data Science for Public Good -
In the summer of 2017, I was a graduate fellow in the Social and Decision Analytics Lab, formerly part of the Biocomplexity Institute at Virginia Tech. It is now the Social and Decision Analytics Division of the Biocomplexity Institute and Initiative at the University of Virginia.
During the summer, I worked on a large variety of projects, which included crime, open-source software, business innovation, structure fires, text data, and hidden populations such as refugees. The results from two of these projects were presented at international conferences in 2018, including in Germany and Switzerland. I also applied many statistical techniques, including spatial and temporal models, network models, and text and sentiment analysis algorithms. I became quite proficient in web-scraping, through APIs as well as customized solutions when APIs weren’t available.