American Inequality is a publication that tracks key factors influencing opportunity in communities across the United States. We highlight inequality in areas such as education, housing, social services and health. We have been featured in the NY Times, NPR, TIME, BBC and more.I lead data science initiatives to identify struggling areas, build predictive models, and create visualizations to share with policy leaders and improve their communities.

• Support Medicare & Retirement CAHPS stars quality programs through data-driven initiatives aimed at enhancing member experience via survey data• Performed statistical tests & analysis to measure the effectiveness of campaigns, evaluating key performance indicators and driving recommendations for campaign optimization• Apply advanced statistics, A/B testing, and machine learning techniques to inform strategic decision-making to identify distressed members• Implement Natural Language Processing algorithms to predict member sentiment and distress based on analysis of unstructured text data, enhancing member engagement strategies• Lead data discovery, analysis, & ingestion processes into Hive database, ensuring seamless integration & accessibility to inform strategic decision-making• Support model development by identifying leading indicators & important features• Engineered backend SQL pipelines to build master data table for survey data, orchestrating data refreshes from several sources, responsible for debugging and providing support for inquiries

• UnitedHealthcare Finance, Dual Special Needs Plans (DSNP) FP&A• Leveraged complex financial data to develop and refine financial models for forecasting and financial analysis• Utilized healthcare metrics including star ratings, risk adjustment analysis, and MLR reporting to drive insights into financial performance• Conducted advanced data querying and manipulation using SQL & SAS on a daily basis for financial reporting• Primary resource for analysis requested by market CFO’s and any corporate level requests for comprehensive data analysis