Projects

Led a team to design, build, test, and deploy an award-winning, web-based facial recognition and celebrity lookalike application called StarMatch. StarMatch compared a user-uploaded  face to thousands of celebrity images and finds the closest lookalike using Convolutional Neural Networks. See demo here.

Aggregated, cleaned, and formatted World Health Organization data on health care expense and non-communicable diseases. Methods used: clustering, decision trees, support vector machines, naive Bayes, knn, and neural networks. Achieved accuracy up to 70% in predicting level of healthcare costs globally.

This project analyzed retail sales data across a number of company products, stores, and locations. Tools used: R Statistical Package, Tableau Visualization Package from Salesforce. This project ranked first in peer cohorts.

Created an application to monitor real time production tank inventories and truck shipments to identify and prevent potential product and revenue loss.

Built a system to monitor and alert for important lease contractual and regulatory dates.

Updated operational dashboards to support natural gas control and dispatch teams.

Built a product recommendation system for liquor stores based on Iowa liquor sales data. Modeled data to find associated products and created an application that takes user input of a specific product and produces the top 10 recommendations of associated products based on sales.

Performed analytical work for a plumbing supply company. Created interactive dashboards for the company to identify potential opportunities to increase profit margins..

Implemented ChiroTouch, a new health records and practice management software system, and trained four fellow employees on how to use it.