 Converted semi-structured congressional yellow pages (1973 – present) into structured data for large-scale analysis, enabling historical research and trend analysis. Trained object detection and OCR models in custom machine learning pipelines for document analysis. Processed quarterly releases of several hundred-page documents, enhancing data extraction accuracy and efficiency through iterative model improvement. Collaborated with interdisciplinary teams to refine document analysis techniques and improve data quality for political science research.

 Collaborate with a team of graduate and undergraduate students on a social avoidance analysis project, focusing on interactions between men and women. Manually label a dataset of several thousand images to train models for social avoidance using self-driving vehicle data, including LiDAR distance measures and bounding boxes for individuals. Contribute to the development of machine learning models capable of analyzing social behavior.

 Wrote and integrated software in .NET and React to train an OCR model integral to instrument workflow. Developed and presented feature design documents, outlining software specifications and requirements. Conducted verification testing and wrote test suites to ensure software reliability for medical instruments. Collaborated with cross-functional teams and customers to align software development with strict FDA compliance.

 Created a data analysis environment using scheduled polling of the Reddit API for posts regarding the maternal Tdap vaccine. Maintained a curated graph database using ArangoDB of over 5,000 documents. Presented at weekly meetings to share my findings with an interdisciplinary research team.

 Crafted dynamic educational materials to improve students’ understanding of course material. Provided both in-person and virtual instruction to over 250 students.