Jan 2021 – Aug 2022 · 1 yr 7 mos · Gurugram, Haryana, India

• Optum Store pricing
• Built a web scraping tool to enable competitive pricing across catalogue
• United Healthcare Digital
• Created a Rshiny call obviation tool to identify and represent the key drivers that drive digital engagement leading to significant reduction in calls made to call centers

Dec 2018 – Jan 2021 · 2 yrs 1 mo · Gurugram

• 1. Slotted Delivery
• Developed the algorithm to identify preferred delivery slots for addresses using historic behavior
• Weekday/Weekend preferences
• 2. Deduplication of warehouse addresses leading to improvement in the service level
• 3.Demand forecast at delivery centers across major sales leading to improved workforce management

Oct 2017 – Dec 2018 · 1 yr 2 mos · Bengaluru, Karnataka, India

• WorkForce Management (Retail, Pharma):
• Led a team of 4 to build ARIMA forecasting model to forecast sales for store clusters enabling optimized workforce management (Time Series Forecasting)
• Designed a Decision board in Tableau to effectively capture anomalies in calls made to inbound IVR leading to proactive decisions(Tableau Reporting)
• Profiled patients more likely to respond and accept campaigns leading to efficient targeting and reduced costs(CHAID Decision Trees).
• Customer Service Operations (Technology):
• Designed a Decision board to capture the failure of different segments of Customer Support Website leading to better customer experience and reduction in the self-failure rate of Articles/Drivers/Diagnostic Engine on the support webpage(R shiny Dashboarding).
• Predicted customers having a high propensity to fail while navigating Customer Support website and accordingly redirecting them to better help content (XgBoost).

Oct 2015 – Sep 2017 · 1 yr 11 mos · Bengaluru, Karnataka, India

• Merchandising Operations (Retail):
• Identified potential stores where a particular item should be introduced from amongst the stores where it is currently not selling (Logistic Regression)
• Defined appropriate metrics and designed methodology to identify a group of substitutable items within a category (Food) based on customer’s shopping behavior (Agglomerative hierarchical clustering).
• Identified the key metric defining substitutability for General Merchandise category and created communities of substitutable items (Ensemble Community Detection).
• Automatic tagging of new products in General Merchandise categories using image recognition (Convolution Neural Networks)