Expert Data Scientist, Fraud Risk Detection

This is a fully remote position, open to applicants in United States.

📋 Description

• Conduct investigations on extensive datasets through exploratory analysis and the development of fraud labels to uncover fraud patterns, attack methods, and behavioral signals.

• Convert unclear fraud and risk challenges into hypotheses, analytical strategies, model specifications, and quantifiable success metrics.

• Create machine learning models aimed at detecting fraud related to account opening, account takeover, and identity risks.

• Assess models using ROC/AUC/KS/Gini metrics, precision/recall, fraud capture rates, false-positive rates, customer friction, and the amount of fraud losses mitigated.

• Develop and authenticate predictive features utilizing identity data, transactional information, consumer credit history, device details, behavioral patterns, temporal data, velocity, network, and third-party resources.

• Write clean, efficient, and thoroughly tested code in Python and PySpark.

• Collaborate with cross-functional teams to implement models and features in batch processing, retrospective analysis, or real-time decision-making environments.

• Supervise feature integrity, model efficacy, population variations, and shifts in fraud patterns.

• Design and deliver analyses that encompass model behavior, trade-offs, risks, and strategic recommendations.

• Adhere to standards concerning data privacy, model documentation, explainability, validation, and governance protocols.

• Collaborate with senior data scientists and engineers from the initial problem definition stage through feature engineering, experimentation, and deployment.

• Report directly to the Senior Manager of Fraud Analytics.

⛳️ Requirements

• Minimum of 5 years of experience in data science, machine learning, statistical modeling, or a relevant quantitative field.

• Bachelor’s or advanced degree in computer science, statistics, engineering, data science, or another quantitative area.

• Proven experience in developing models for fraud detection, identity risk, credit risk, financial crime, or other adversarial risk assessments.

• Demonstrated ability to create significant fraud-related features.

• Proficiency in Python and PySpark programming languages.

• Experience in writing modular and tested code suitable for managing large datasets and distributed or cloud-based data systems.

• Familiarity with technologies such as pandas, NumPy, scikit-learn, XGBoost, TensorFlow, or similar tools.

• Understanding of supervised learning, model evaluation techniques, feature selection, statistical inference, experimentation, and model calibration.

• Experience dealing with class imbalances, delayed or incomplete labels, evolving attack patterns, and model drift.

• Experience in transitioning models to production either directly or in close collaboration with engineering teams.

🏝️ Benefits

• Competitive compensation package along with a bonus plan.

• Comprehensive medical, dental, and vision benefits.

• Matching 401(k) contributions.

• Flexible work arrangements, offering remote, hybrid, or in-office options.

• Flexible time off policies including volunteer time off, vacation, sick leave, and 12 paid holidays.

• Variable pay opportunities.

• Extensive benefits package.