Staff Data Scientist Fraud and Risk

Serve as a Staff Data Scientist in Fraud and Risk, building and improving classical machine-learning models for global fraud detection and risk mitigation. Develop production GenAI and agentic workflows, manage drift and imbalanced data, guide technical standards, and partner with engineering, product, and business stakeholders.

Responsibilities

• Design and develop predictive machine-learning models for fraud detection, risk assessment, and anomaly detection.
• Design and develop production-grade GenAI and agentic AI solutions.
• Handle highly imbalanced datasets.
• Identify, measure, and resolve data drift and concept drift.
• Present insights and translate model outputs for non-technical stakeholders.
• Set technical standards and review architectures.
• Collaborate with ML engineers to deploy models.
• Establish CI/CD pipelines and implement model tracking and observability.
• Partner with Data Engineering to optimize feature engineering and feature stores.
• Translate fraud typologies and business requirements into data-science problems.
• Build and track model performance metrics and business impact.
• Propose innovative solutions in ambiguous situations.

Requirements

• 10+ years of experience in Data Science or Machine Learning.
• Substantial experience fighting fraud or mitigating risk.
• Deep theoretical and practical understanding of classical machine-learning algorithms.
• Experience with XGBoost, LightGBM, Random Forests, SVMs, and ensemble methods.
• Experience building adversarial fraud-detection models.
• Experience with synthetic data generation.
• Production experience with GenAI and multi-agent systems.
• Experience with LangGraph, AWS Bedrock, or the GCP ecosystem.
• Experience handling highly imbalanced datasets.
• Knowledge of sampling methods, cost-sensitive learning, and evaluation metrics.
• Experience detecting and addressing data drift, concept drift, and model degradation.
• Strong SQL and Python skills.
• Exposure to PyTorch and Hugging Face.
• Knowledge of ML and data-science inference performance optimization.
• Understanding of Spark and Hadoop.
• Strong knowledge of AWS or GCP.
• Experience with Databricks, Snowflake, BigQuery, Tecton, or Fiddler.

Benefits

• Competitive salary and share options
• Generous annual leave
• Paid maternity, paternity, and adoption leave
• Sabbatical leave options
• Private family health insurance
• Therapy sessions, courses, meditations, and workshops
• Paid volunteering and development days
• Annual learning and development budget
• Work from abroad for up to 90 days annually
• Home office setup contribution
• Laptop replacement benefit
• Office snacks, coffee, tea, and lunch