Lead Fraud Data Scientist
Core
Lead a fraud detection team to design, build, and deploy real-time machine learning models that protect the company and customers from fraudulent activity in cross-border money movement and lending.
Role type
Senior IC Lead Data Scientist (Fraud)
Builds
Production ML models for fraud detection, credit/lending risk mitigation, and real-time risk scoring.
Domain
Fintech, cross-border payments, consumer lending, and risk management.
Deliverable
production ML models
Required skills
Python (pandas, scikit-learn), SQL, tree-based ML models (XGBoost, CatBoost, LightGBM), statistical models (Logistic Regression, Lasso/Ridge), model explainability (SHAP, LIME), sampling techniques for imbalanced data, unsupervised learning (clustering, outlier detection), cloud platform deployment (GCP, AWS, Azure), A/B test design, advanced English.
Preferred skills
FinTech/risk/fraud domain experience, alternative/bureau data experience, Graph Neural Networks (GNNs), regulatory familiarity (FCRA, ECOA), MLOps, GCP Vertex AI, Spanish/Portuguese.
Technologies
Python, pandas, scikit-learn, XGBoost, CatBoost, LightGBM, SHAP, LIME, Tableau, Looker, GCP, AWS, Azure, Neo4j, NetworkX, Vertex AI.
Responsibilities
Define long-term ML strategy and mentor junior data scientists; own end-to-end model lifecycle from exploration to monitoring; design models for synthetic identity and application fraud; conduct deep-dive investigations using clustering and network analysis; execute A/B tests to measure model impact; collaborate with Product, Engineering, and Risk teams; deploy and maintain models in cloud environments; build dashboards to track fraud loss rates and false positive rates.
Seniority
Senior, hands-on IC with leadership responsibilities
