Senior Data Scientist - Fraud Risk Analytics - Fintech

About the job:

About the Role:

We are looking for a Senior Data Scientist to support fraud strategy and analytics across a consumer lending portfolio.

You will analyze fraud trends, monitor decisioning outcomes, and recommend practical changes to policies, rules, thresholds, and reporting.

This is a hands-on fraud-risk analytics and strategy role, not a role primarily focused on building machine learning models.

You will partner with fraud operations, product, credit/risk, data engineering, and external vendors to identify emerging risks and improving fraud controls.

Key Responsibilities:

- Monitor fraud performance, including losses, capture rates, false positives, approval impact, vintage trends, and segment-level KPIs.

- Investigate emerging fraud patterns using application, account, behavioral, device, transaction, and third-party data. Quantify their impact and recommend mitigations.

- Translate fraud analysis and investigator findings into recommendations for decision-engine rules, thresholds, policies, and operational processes.

- Monitor fraud scores and scorecards for drift or deterioration and recommend adjustments or escalation when needed.

- Analyze policy backtests and champion/challenger tests to assess the impact of strategy changes on fraud outcomes, approvals, and downstream credit performance.

- Evaluate third-party fraud and identity signals and recommend how they should be used in fraud decisioning.

- Produce recurring fraud reports and executive-level analyses covering loss drivers, fraud typologies, and decisioning outcomes.

- Work with cross-functional teams to implement changes and improve fraud data, workflows, and monitoring.

Required Qualifications:

- 2 - 4 years of hands-on, professional fraud-risk strategy or fraud analytics experience in financial services or fintech.

- Experience investigating fraud patterns in application, account, or transaction data and recommending changes to fraud rules, thresholds, controls, or operational processes.

- Strong SQL skills and practical Python proficiency for fraud analysis, segmentation, automation, and reporting.

- Working knowledge of fraud scores or scorecards, including how to interpret performance and identify when a strategy change is needed. Building ML models is preferred.

- Understanding of fraud typologies such as identity fraud, first-party fraud, synthetic identity fraud, or third-party fraud.

- Ability to explain analytical findings and recommendations to technical and non-technical stakeholders.

- Bachelor's degree in a quantitative field or related discipline.

Preferred Qualifications:

- Experience in consumer lending, loan applications, or other high-fraud-risk credit products.

- Exposure to fraud decision engines and third-party identity, device, or transaction signals.

- Experience assessing fraud-rule changes through backtesting or champion/challenger analysis.

- Familiarity with US consumer lending risk practices or fraud-ring analysis.