Data Scientist
CurrentEnabled self-service extraction of actionable insight from FDA medical device 510(k) clearance summaries for medical device investors and creators.Created automated ETL pipeline - data mining, web scrapingCreated anomaly detection and validation procedures to ensure corpus data quality.Created a recommender system to improve predicate discovery.Managed Solr index, implemented faceted result sorting and computed fields. Proposed, prototyped and implemented push notification of new results for user-saved queries.Optimized service availability using metrics-based assessmentsImproving predicate discovery using machine learning, natural language processing, topic modeling and automated summarization.