: Your Way to Impact In this role, you will own an end-to-end portfolio of PayPal products or markets, driving accountability for loss and decline rates while managing a small team of Data Scientists (3–6).
You'll collaborate with multiple stakeholders to design holistic fraud-prevention strategies, from initial data exploration through model deployment and performance monitoring.
By striking a balance between speed and quality, you'll ensure that our fraud-risk function makes a meaningful contribution to business growth, meeting key KPIs such as authentication rate targets and overall loss-rate objectives.
Day-to-Day Responsibilities Portfolio Ownership: Lead a group of Data Scientists to manage fraud and loss metrics for assigned product lines or markets.
Strategy Development: Collaborate cross-functionally to develop and iterate on fraud-prevention frameworks that optimize transaction declines and minimize customer friction.
Model Lifecycle Management: Guide the team in data preparation, feature engineering, model training, and validation—leveraging Python/R and cloud-native platforms (e.g., BigQuery, Snowflake).
Stakeholder Alignment: Host regular syncs with Business Units, Risk Operations, and Compliance to ensure models and business rules remain aligned with evolving policy and regulatory requirements.
Performance Monitoring: Track model health (drift, false-positive/false-negative rates) and lead root-cause analyses for any anomalies or spikes in loss.
What You'll Bring 8+ years of relevant experience working with large-scale, complex datasets, including at least 3 years managing a team of 3–6 Data Scientists in a fraud-risk or financial-services environment.
Proven ability to decompose ambiguous business requirements into a structured analytic plan and deliver data-driven recommendations.
Advanced proficiency in SQL and Python (or R) for data wrangling, EDA, and model development.
Expertise in exploratory data analysis and preparing clean, structured datasets for modelling Hands-on experience applying supervised and unsupervised learning techniques (regression, classification, clustering, decision trees, anomaly detection).
Familiarity with production ML frameworks (scikit-learn, TensorFlow, PyTorch) and cloud data platforms (BigQuery, Snowflake).
Deep understanding of fraud-risk principles, including AML/KYC, regulatory compliance, and performance metrics (Precision, Recall, ROC-AUC).
Strong track record of partnering with engineers, product managers, and business leaders.
Excellent verbal and written communication skills, able to distill complex findings into clear narratives for both technical and non-technical audiences.
A passion for inventing new approaches to big, ambiguous problems—always looking to build novel solutions that stay ahead of evolving threat vectors.
Additional Responsibilities