• Coded & deployed 5 R Shiny interactive dashboard for tracking forecasting variance and demonstrating top 10 medical type drivers boosting paid amount in marketplace (32 states) for generic, brand, and specialty medications; boosted efficiency by 76.8% for departments and saved 40+ hours per month.

• Managed RDBMS&EDW database using MySQL and ETL from Teradata to support Finance, risk management and health care departments, and built dynamic script to monitor and validate the process of KPI reporting system migration, improving the validation efficiency by 85%.

• Advanced forecasting modeling using Python healthcare operation, increased forecasting accuracy MoM by 34.5% and redesigned customer segments and performed extensive A/B testing to decrease in script per member by 21.3% and cost per script by 29.7%.

• Presented healthcare insights using PowerBI to visualize elevating issues, outlier trends and management operational impacts and recommended required business strategies by simulating model, leading to a 17.5% reduction in PMPM (per member per month) cost and 24.7% increase in MM (member month) utilizer index bi-weekly.

Client: Global Multi-Brand Fitness Company

• Built a predictive model using random forest based on engagement activities which reduced customer churn by15% due to early-stage intervention.

• Identified significant features that improved personal training programs by using segmentation analysis which resulted in a 5% increase in personal training sessions.

• Applied A/B testing to identify the most effective communication strategies which resulted in 3% increase in customers.

Client: Leading Hospitality and Entertainment Business

• Performed exploratory analysis using Python and Tableau to gain a comprehensive understanding of current promotional strategies and promotional offers influencing customer trip decisions.

• Implemented k-means clustering algorithm to segment and profile 50M consumer data set to understand promotional recommendations, which led to an increase in new customers by 6% and improved the coupon redemption rate by 8%.

Client: Mall of America

• Segmented 3 years of customer service calls using K-means algorithm in R to identify latent patterns and improper resource spends; presented the recommendations using Tableau.

• Optimized staffing by analyzing factors influencing call volume using regression analysis; utilized time series to forecast based on 40K+ calls in Python to achieve a 10% decrease in redundant resource utilization.