Data Scientist (Fraud)

Role Overview

You will be the quantitative backbone of our fraud-fighting efforts, working directly on the models and detection systems that safeguard our customers and merchants. Day to day, this means digging into transaction data to uncover abuse patterns, prototyping and refining machine learning solutions, and partnering with engineering and product teams to move those solutions from notebook to production. Your work will directly influence how we stay ahead of fraudsters, so a mix of technical rigour and commercial instinct is essential.

Key Responsibilities

• Develop and evaluate machine learning models for fraud detection, then take ownership of their performance in production through regular monitoring, retraining, and iteration.
• Design controlled experiments that measure the real-world impact of fraud interventions, balancing friction for genuine customers against reductions in financial loss.
• Quantify the scale and impact of different fraud typologies across our product portfolio, producing insights that guide where we invest time and resources.
• Build anomaly detection tools and heuristics that flag novel or evolving fraud patterns early, before they turn into large-scale attacks.
• Partner with fraud operations, software engineers, product managers, and analysts to convert model outputs into clear, actionable mitigation workflows.
• Investigate unusual patterns in data to identify emerging fraud vectors and translate those findings into features or model improvements.
• Communicate technical trade-offs and experiment results to non-technical stakeholders, helping them make informed decisions about fraud strategy.

Requirements & Qualifications

• Strong statistical foundations, backed by a degree in statistics, mathematics, engineering, computer science, or another quantitative discipline.
• At least 3 years of experience in data science, decision science, risk analytics, or a related field, ideally within fraud, payments, or financial crime.
• Proven track record of building and shipping machine learning models in a production environment, not just in ad-hoc analyses.
• Solid grasp of core data science skills: experimental design, statistical inference, model evaluation, and feature engineering.
• High-level proficiency in Python and SQL, with comfort working across the entire model lifecycle, from data exploration to deployment and monitoring.
• An investigative mindset — you naturally look for patterns, anomalies, and explanations that others might overlook.
• Strong communication skills, able to present complex findings clearly to both technical and non-technical audiences, and to turn insights into practical recommendations.
• Experience in fraud, risk, or financial services is a significant advantage, though we welcome candidates with transferable analytical depth from adjacent domains.

What We Offer / Why Join

This is a high-ownership role where your models and analyses will have a direct, visible impact on protecting millions of users. You will work in a fast-paced environment alongside fraud operations, engineers, and product leaders who are genuinely invested in data-driven decision-making. We offer the opportunity to tackle some of the most challenging problems in financial crime while seeing your work scale across a real production platform.