Data Science Work • Developed Semantic Enrichment Service (SES) from base Flair named entity recognition (NER) tagger via Spacy NLP pipeline with other data scientists. • Evaluated performance of the service using precision/recall metrics. • Tested and implemented best NER tagger, including asynchronous Azure Cognitive Services, to replace Flair based on accuracy and performance metrics. • Hackathon (4th place) - Developed a subjectivity SVM analysis on news data as a feature of a new product that presents news. Other features included news summarization and news classification. • NetOwl software - Wrote NetOwl entity extractor rules to extract and distinguish entity types like company, facility, product. Improved precision/recall values of company entity types by 10%.Software Development Process Work • Built and deployed SES pipeline to FastApi in Jenkins and tested production performance on Locust. • Created custom utilities python package from methods used in SES and uploaded to JFrog via Pypi. • Maintained SES and custom packaging code by adhering to Black python code format standards, writing unit tests, and measuring coverage with SonarQube. Versioned SES code with Git. • Versioned NetOwl code with ClearCase, and deployed code to HPCC (computing cluster) with Tarball files • Used Confluence for documentation.

Improved mathematical models for Kitware on the IBAMR project. Member of a team enhancing Kitware’s Computational Model Builder. Wrote XML for GUI and coded and tested Python for CMB’s export script for IBAMR simulation. (Boyce Griffith, PI)

Developed software for factor analysis in R for independent research using Dr. Eric Youngstrom’s survey results. Wrote R code to tidy the survey results to analyze specific datasets. Advised the professor how to code efficiently. Results from my factor analysis were displayed at the NCPA conference.