Created Signal, a scalable predictive modeling software that prospectively identifies clinical, financial, and patient engagement risks for members impacted by substance use disorder. Signal supports operations and clinical teams through risk management and patient prioritization.Tools: R, SQL, Python, AWS, Git, Docker

Developed MedRec, a R Shiny app supporting 100+ users performing patient medication reconciliations

Purpose: Scale the Data Science footprint by engaging customers and stakeholders to develop integrated analytic products.Technologies: R, Python, SQL, AWS, Docker, R Shiny, GitCurrent project: Scaled the Data Science footprint by listening to customer feedback, deploying a modern solution portfolio using AWS services, and developing training content. Fostered relationships with stakeholders in technical and business areas to discover use cases and drive adoption. Launched 6 medium to large-scale internal analytics products and services including predictive models, web applications, and APIs. Authored over 100 pages of documentation containing reproducible code snippets. Mentored 3 current or aspiring Data Scientists.

Purpose: Support Sales and Marketing with analytics leadership to optimize business processes, research business opportunities, and drive strategy.Technologies: R, SQL, Tableau, AWSHighlight projects: Imagined and developed an innovative text classification algorithm to enable 100% coverage of marketing campaign data audits. Classifier prediction accuracy at 99%. Significantly improves customer experiences by discovering and eliminating data errors in large mailing lists. Deployed as a web application on AWS and utilizes C++ and parallel processing for fast performance.

Purpose: Improve planning and operations using analytics and predictive modeling.Technologies: R, SQLHighlight project: Developed a model to predict Medicare Star measure thresholds using linear regression, ARIMA forecasting. and hierarchical clustering. Contributed to $30M in revenue. Patent pending.

Purpose: Support senior researchers through data mining, analysis, and interactive visualization.Technologies: RHighlight project: Hierarchical clustering applied to US states for the purpose of segmentation based on variables related to corruption.