Senior Data Scientist – Fraud Model Validation

Overview

In this second-line role, you validate and stress-test fraud ML models to ensure reliability before deployment at scale. You’ll work closely with first-line teams to challenge methodologies, reproduce results, and surface risks across data, features, and production pipelines. Your work helps Klarna manage fraud risk across high-volume transactions, leveraging diverse models and emerging techniques. You’ll shape validation tooling and governance to keep models trustworthy as they evolve with business needs.

Responsibilities

• Assess model performance using fraud-specific metrics and business trade-offs
• Review large transaction datasets and feature pipelines for representativeness and leakage risk
• Evaluate drift detection, retraining strategies, and production monitoring practices
• Assess CI/CD and deployment controls across Docker, Jenkins, and AWS environments
• Review model governance, explainability, and compliance with regulatory expectations on model risk and privacy
• Validate emerging techniques (graph networks, anomaly detection, GenAI-based systems) and document validation outcomes
• Communicate model risks to data scientists, ML engineers, and business stakeholders

Key requirements

• 3+ years hands-on fraud-related modeling
• Proficiency in Python and SQL; experience with PySpark/Spark
• Strong knowledge of tree-based models (e.g., LightGBM) and anomaly detection
• Experience with full ML lifecycle: feature engineering, deployment, monitoring
• Understanding of model risk governance, bias, fairness, and privacy risk
• Ability to explain complex models to technical and non-technical audiences
• Experience with agentic AI workflows and automation

• clear communication
• analytical thinking
• stakeholder management
• LightGBM
• scikit-learn
• graph or network models

…