Nicole Roberts Email & Phone Number

Bengaluru, Karnataka, Ind..., Karnataka, IN

Kansas City, Missouri, US

• Research and build machine learning models for various applications within Marketing, with a focus on customer segmentation and lead management for retail and intermediary markets.
• Use predictive analytics to help determine where Marketing budget is best optimized, including Marketing Mix Models, LSTM neural networks, and tree ensemble methods.
• Spearhead marketing data analysis and model building for American Century’s partner brand, Avantis, assisting in the overall marketing roadmap for this relatively new asset management firm.
• Work closely with stakeholders to identify, design, and implement internal process improvements including automating manual processes, optimizing data delivery, and re-designing infrastructure.

Columbus, OH, US

• Worked closely with Marketing, Finance, and Allocation teams to understand business needs and develop innovative strategies and operational improvements through data-driven methods.
• Performed extensive feature engineering using SQL for machine learning models based on a thorough understanding of retail customer behavior and end-goal use cases.
• Developed a two-part machine learning model that predicts what discount amount a member generally requires to make a purchase. When applied to Marketing’s campaign outreach, model boosts revenue by eight million dollars.
• Built and automated a Customer Segmentation Map based on a member’s monetary and frequency shopping patterns that feeds into a machine learning model predicting customer segment migration patterns.
• Increased revenue by four million dollars by isolating and excluding from Marketing campaign offers customers with a high return rate and low customer lifetime value score.

New York, NY, US

• Developed a proprietary machine learning pipeline using a set of option contract variables as features along with, notably, predicted VIX prices (derived from own ‘Modeling the VIX with LSTM’ project model).
• Provide traders confidence in their investment decisions, accurately predicting ROI outcome with up to 77% accuracy and 85% precision.
• Optimize model parameters with ongoing back testing.
• Develop additional predictive technical indicators to identify optimal entries and exits.

NEW YORK, NY, US

• Simple the App is a cloud-based field service management SaaS solution that streamlines internal business processes through both a mobile app and desktop dashboard.
• Architected and programmed CRM system via a Flask web framework and Jinja2 template engine with PostgreSQL, including real-time scheduling and dispatching through an automated app-to-technician SMS text messaging.
• Drastically reduced the time it takes companies to assign and manage workload and maintain organized, up-to-date accounting; beta version testing thus far implies the following metrics: - Freed up ~82% of dispatcher’s.