For the past 5 years, I've been responsible for developing, organizing, and training the team responsible for fraud detection and prevention at Mountain America Credit Union. My team consists of 18 individuals, which includes Data Analysts/Data Scientists, Data Engineers, and Fraud Professionals. Collectively, we prevent approximately $30 million in fraud per year. I led an effort to develop several in-house detection systems including an alert system, facial recognition, check image risk profiling, and traditional structured data such as transaction monitoring.

Developed model that saves $100k per month in check fraud. (Currently in production) Developing a model (currently in testing) that should 3X these monthly savings. I serve as department SQL expert. I use this expertise to mine transaction histories for approximately 700,000 members (tens of millions of transactions), and identify emerging fraud trends. Developed bayesian network model to identify common points of compromise for related card fraud cases. To date this algorithm has successfully identified 3 common points of compromise before public announcement of data breach (confirmed hits) with additional unconfirmed compromises identified saving members thousands of dollars in additional fraud losses.Built a custom machine learning models for expert system rule optimization to both minimize criteria (reducing complexity thus increasing maintainability of the system) , and maximize fraud caught saving thousands every month.

A/B testing for online and mobile websites to optimize engagement with banking sites. Analyzed data in regards to mobile app to optimize usage working closely with UX designers.Analyzed Digital Marketing Analytics for financial products to increase revenue, and find problems with digital marketing campaigns.Helped to make important strategic decisions about online and mobile products. Deciding when to retire products, introduce new products, and revamp other products.Helped to develop a recommendation engine for financial wellness that served recommendations to approximately 300,000 members per month. (About 3 million recommendations per month total)Developed algorithms that helped in decision making for high-risk mobile loans (loans competing with payday lenders).Developed automated solutions that improved data processing, and reporting reducing processing time for mobile-originated loans from 1 week to under 24 hours.Served as expert for google analytics and adobe analytics within department.

Analytical Methods in Applied Economics (APEC 3310)Introduction to Natural Resource and Regional Economics (APEC 3012) ONLINE