As an Applied AI Machine Learning VP (Fraud Modelling) in the ICB Risk Modelling team, you will play a crucial role in developing and managing machine learning models used to mitigate fraud risk within ICB.
You will work with multiple partner teams—including Strategy, Technology, Product Management, Legal, Compliance, Business Management, and Model Governance — to ensure the models meet the firm's high governance standards and regulatory requirements, and support audit and other business functions around model management.
The primary focus will be on identity verification fraud, where you will lead efforts to adopt and implement advanced solutions, including models from leading vendors for detecting and preventing fraudulent activities.
Job Responsibilities
- Develop and manage proprietary fraud models and perform due diligence for vendor models. Validate model performance on internal data, ensure modelling choices are appropriate for the portfolio, decisioning context, and operational constraints. Work closely with platform engineers to support model deployment.
- Prepare complete governance and Model Risk Management packages, including development documentation, testing evidence, model limitations, and implementation specifications. Support independent validation, respond to findings, and drive remediation to closure.
- Support ongoing monitoring for performance and stability (e.g., drift, calibration, population shifts, fraud-typology changes). Define monitoring metrics, thresholds, Investigate degradations and drive remediation actions.
- Communicate model design, trade-offs, results, and limitations to senior stakeholders and governance committees in clear business terms. Train and support downstream users on correct interpretation and use of model outputs.
- Maintain audit-ready artifacts such as model documentation, monitoring reports, validation responses, and control evidence to support internal audits and regulatory exams. Provide timely, traceable responses to inquiries and ensure documentation stays current post-deployment.
Required Qualifications, Capabilities, and Skills
- Advanced degree (MSc or PhD) in a quantitative or technical discipline.
- Solid understanding of fraud modelling in financial organizations, including the unique challenges and regulatory considerations involved. Credit modelling is acceptable as a transferable background.
- Industry experience in applied data science, machine learning techniques, with a strong understanding of both traditional statistical and machine learning models.
- Proficient in Python, SQL, with hands-on experience in data analysis and writing production-quality code. Extensive experience with machine learning and data analysis toolkits (e.g., NumPy, Scikit-Learn, Pandas).
- Ability to effectively leverage Generative AI tools to enhance productivity, analysis, and problem-solving in day-to-day work.
-
Strong written and spoken communication skills to effectively convey technical concepts and results to both technical and business audiences. Team player.
Preferred Qualifications, Capabilities, and Skills
- Experience with identity verification fraud models.
- Experience with ML model explainability.
- Experience with model risk management frameworks.