Data Scientist

Primary Purpose of Job

Responsible for working with large and diverse data sets (i.e. data from dozens of sources including transactional, online, device, financial, and other sources) and utilizing machine learning and statistical methods to identify fraud patterns and develop strategies and rules for managing fraud. Leads the design, development, and execution of fraud and BSA/AML strategies that will deliver a great customer experience while keeping losses manageable.

Major Job Accountabilities

1. Discovers insights from various data sources that can help predict fraud and lead development of fraud strategies, rules analysis and predictive models to manage fraud while keeping a balance with the customer experience. For example: Identifying transactional anomalies, designing and executing specialized queries that can help streamline compliance processes and reduce "false positive" alerts, providing teammates more time to review more-targeted cases of suspicious activity.
2. Develops and manages a framework to evaluate the efficacy of existing rules and models to reduce financial exposure in both the fraud and BSA/AML frameworks. Builds analytic models using a variety of techniques such as logistic regression, risk scorecards, and pattern recognition technologies.
3. Collaborates with 1) Fraud prevention/detection teams and Operations to understand business needs, data generating process, system capability, and potential impact of models; 2) analytics partners in credit risk and marketing to leverage their knowledge and expertise in developing fraud strategies and 3) Compliance teammates in support of HMDA and other data analytics needs.
4. Analyzes large quantities of data to identify fraud patterns and loss manifestation.
5. Performs model and data validation testing for departmental models in support of the Model Risk Management framework.
6. Develops, produces, and maintains standard and ad hoc reports and presentations for the Fraud team to monitor trends and drive efficiency improvements.
7. Collaborates with business lines on strategic projects that enhance fraud prevention and financial crime detection.
8. Provides indirect leadership of individuals and groups on a near daily basis, e.g., recommending process improvements or enhancements to those in other functional areas.

Experience Required

Minimum five (5) years of experience in the following:

1. Working experience in analytics or data science type role.
2. Experience in financial services required.
3. Fraud/BSA-AML/Compliance analytics a plus.

Minimum three (3) years of experience in the following:

1. Experience in model development/machine learning like logistic regression, random forest, boosting, neural networks and decision trees.
2. Other machine learning methods a plus.

Required Skills or Training

1. Demonstrates an effective project management or comparable organizational skill set, including ability to isolate causal factors and develop plans for improving processes.
2. Direct experience on development of models, systematically processing big data and quality assurance work on the final output.
3. Excellent written and oral communication skills.
4. Experience building predictive models within financial services or insurance industries is a plus.
5. Experience in developing fraud detection/prevention models in financial industry is a plus.
6. Experience with experimental designs and data visualization tools.
7. Experience working with credit card, debit card, payment fraud, bust-out, first party fraud or AML risk modeling is a plus.
8. Excellent SQL/relational database skills are required.
9. Excellent statistical and machine learning skills with at least one of the programming tools (e.g., Python, R or similar).

Professional Certifications, Licenses, And/or Registration Requirements

Certified Fraud Examiner (CFE) or Association of Certified Anti-Money Laundering Specialists (ACAMS) equivalent certifications is a plus.