Senior Data Scientist - Fraud Model Validation (Fintech)

Location: London, United Kingdom; hybrid, with most teams meeting in the office 2–3 days per week

Salary: £91,196–£116,438 per year

Company

Klarna builds an everyday finance network serving consumers across multiple countries, using AI to develop financial products and services.

What you will do

• Independently validate fraud models covering payments, logins, identity, and account security.
• Reproduce model results, build challenger models, and assess performance using precision/recall, ROC-AUC, PR-AUC, cost-sensitive metrics, and fraud capture rate.
• Review transaction datasets exceeding 100 million records and feature pipelines with hundreds of features for representativeness, leakage, and bias.
• Evaluate drift detection, retraining strategies, monitoring, CI/CD, and deployment controls across Docker, Jenkins, and AWS SageMaker, S3, Athena, and Lambda.
• Validate graph networks, behavioral biometrics, anomaly detection, and GenAI-based systems.
• Document findings and communicate model risks to data scientists, ML engineers, and business stakeholders.

Requirements

• 3+ years of hands-on experience with fraud-related modeling, including transaction, account takeover, identity, or payments fraud.
• Strong knowledge of LightGBM, anomaly detection, graph or network models, and full-lifecycle ML development.
• Fluency in Python and SQL, with experience using PySpark or Spark for large-scale data processing.
• Experience designing agentic AI workflows and automation systems.
• Knowledge of model validation, risk governance, bias, fairness, explainability, data privacy, and regulatory expectations.
• Ability to work in a hybrid setup in London, with regular office attendance.

Nice to have

• Master’s or PhD in data science, statistics, mathematics, computer science, physics, engineering, or another quantitative field.
• Experience with BNPL, credit cards, or other transaction-heavy payment products.
• Experience mentoring junior validators or leading validation reviews.
• Experience with rejected-transaction inference, AI governance frameworks, or emerging AI regulations.

Culture & Benefits

• Independent second-line validation role working closely with first-line fraud data science and ML engineering teams.
• Co-located collaboration with office attendance typically 2–3 days per week, depending on the team.
• Diverse professional backgrounds and perspectives are welcomed.
• Final compensation depends on qualifications, skills, and experience.

Hiring process

• Submit a CV in English emphasizing concrete outcomes, deliverables, and business impact.