Senior Data Scientist, Fraud Prevention

About Highnote 

Founded in 2020 by a team of leaders from Braintree, PayPal, and Lending Club, Highnote is an embedded finance company that sets the standard in modern card platform management. As an all-in-one card issuer processor and program management platform, we provide digital-first organizations with the flexibility to seamlessly issue and process payment cards, embed virtual and physical card payments, and integrate ledger and wallet functionalities—empowering businesses to drive growth and profitability.

We’ve raised $145M+ and have grown our team to 125+ employees. Headquartered in San Francisco, we’ve managed to build one of the most advanced payments teams in the industry, with team members in 25+ US states. 

Operating through our core values of customer obsession, executional excellence, intentional inclusion, we’re helping businesses grow for the future by creating the payment products demanded by tomorrow, with the ability to solve for use cases that don’t exist yet.

We are fast-moving, hands-on, and strongly believe everyone deserves a seat at the table. We believe we’re unlocking incredible opportunities that can change the future of payments, as long as we have the right people to make it happen. 

Job Description

As a Data Scientist focused on Fraud Detection, you will be tasked with the critical mission of protecting Highnote and our innovative clients from financial losses and reputational damage. The position requires a blend of rigorous analytical skills, deep machine learning expertise, and a growing understanding of the payments fraud domain. Success in this role means effectively identifying and mitigating threats from issuance card transaction fraud and both first-party and third-party account application fraud, thereby ensuring a secure and seamless experience for legitimate users while thwarting malicious actors. 

Key Responsibilities:

1. Data Analysis & Pattern Identification: Conduct thorough exploratory data analysis on large, complex datasets encompassing transaction logs, application details, user behavior patterns, device information, network data, and other relevant sources. Your goal is to proactively identify emerging fraud trends, sophisticated attack patterns, and subtle anomalies specific to card issuance transactions and account application vulnerabilities (including first-party and third-party identity fraud).
2. Model Development & Implementation: Design, develop, rigorously validate, deploy into production, and meticulously monitor machine learning models. These may include techniques like logistic regression, decision trees, random forests, gradient boosting machines, clustering, or anomaly detection (outlier detection) algorithms. Complement these models with well-defined rule-based systems where appropriate. The objective is to detect and prevent fraudulent activities in real-time or near real-time, adapting models as new fraud patterns emerge.4
3. Feature Engineering: Conceptualize, create, and refine impactful features for fraud detection models. This involves leveraging diverse internal data streams from Highnote's unified platform and potentially integrating external data sources. Strategy Optimization & Experimentation: Design, execute, and analyze A/B tests and other controlled experiments to rigorously evaluate the performance of different fraud detection strategies, specific rules, and machine learning models. The focus is on optimizing the trade-off between maximizing fraud prevention effectiveness and minimizing the impact on legitimate user transactions and onboarding experiences (reducing false positives).
4. Performance Monitoring & Iteration: Establish and maintain comprehensive monitoring systems, dashboards, and reporting mechanisms to continuously track model accuracy, rule precision/recall, key fraud metrics (e.g., detection rates, false positive rates), and overall system health.Cross-Functional Collaboration: Work closely and effectively with Engineering, Product Management, Risk Operations, and Compliance teams. This collaboration is essential for understanding business requirements, operationalizing models and rules efficiently within production systems, integrating necessary data pipelines, and providing actionable, data-driven insights to inform broader risk strategies.
5. Communication & Reporting: Clearly and concisely communicate complex analytical findings, model behaviors, performance metrics, and strategic recommendations to diverse audiences, including technical peers, operational teams, product managers, and executive leadership.

Qualifications

• A Bachelor's or Master's degree in a quantitative field such as