Fraud Strategy Data Scientist

About the role

BILL is seeking a Fraud Strategy Data Scientist to join their growing team in San Jose, California. This role focuses on building data-driven models and strategies to detect and prevent fraudulent activity across the BILL payment platform and services. The scientist will work closely with cross-functional teams to translate complex analytical findings into actionable business recommendations that meaningfully reduce financial risk. You will own the end-to-end lifecycle of fraud detection projects, from initial hypothesis formulation through deployment and ongoing performance monitoring and reporting.

Key facts

What you'll do

• Design and implement statistical models to identify fraudulent transaction patterns across BILL payment systems and platforms.
• Collaborate with product and engineering teams to integrate fraud detection models into existing production workflows effectively.
• Analyze large datasets of historical transaction records to uncover anomalies and emerging fraud trends over time.
• Develop real-time scoring algorithms that assign dynamic risk levels to incoming transactions as they are processed.
• Create interactive dashboards and automated reporting tools to communicate fraud metrics to stakeholders and leadership teams.
• Conduct rigorous A/B tests and controlled experiments to measure the effectiveness of fraud prevention strategies.
• Investigate flagged transactions and provide detailed analytical summaries to support dispute resolution and case management.
• Build and maintain machine learning pipelines for continuous model retraining, validation, and performance monitoring in production.
• Partner with compliance and legal teams to ensure all fraud strategies meet current regulatory requirements.
• Document all methodologies, model assumptions, and validation results for both internal review and external audit purposes.
• Present findings and recommendations to senior leadership during regular strategy reviews and cross-departmental meetings.
• Work with data engineering teams to define data quality standards and improve feature engineering for fraud models.
• Evaluate model performance using precision, recall, and other classification metrics to optimize fraud detection accuracy.
• Stay current with industry research and emerging fraud techniques to continuously improve detection capabilities and reduce false positive rates.

Requirements

• Bachelor's degree in a quantitative field such as statistics, mathematics, computer science, or economics.
• At least three years of experience in data science, analytics, or a related technical role.
• Proficiency in SQL for querying and manipulating large relational databases with complex joins.
• Strong experience with Python or R for statistical modeling and data analysis tasks.
• Familiarity with machine learning frameworks such as scikit-learn or XGBoost for classification problems.
• Solid understanding of fraud detection methodologies and common attack vectors in payment systems.
• Demonstrated ability to communicate complex analytical results clearly to non-technical business audiences.
• Experience working with payment processing or financial technology systems in a production environment.
• Experience with feature engineering and dimensionality reduction techniques to improve model efficiency and accuracy.
• Strong written and verbal communication skills with the ability to present findings to diverse audiences.
• Ability to work independently and manage multiple projects with competing priorities in a fast-paced environment.

Nice to have

• Graduate degree in a quantitative or analytical discipline such as a master's or PhD.
• Prior experience working in the financial services or payments industry is highly valued.
• Knowledge of graph-based analytics and network analysis techniques for detecting organized fraud rings.
• Familiarity with cloud-based data platforms such as AWS or Google Cloud Platform is a plus.

Skills & tools

• Python programming for data analysis and machine learning model development.
• SQL for data extraction, transformation, and querying large datasets at scale.
• Using scikit-learn and XGBoost for building and evaluating predictive classification models.
• Tableau or Looker for creating interactive visualizations and fraud detection dashboards.
• Using Jupyter notebooks for exploratory data analysis and rapid prototyping workflows.
• Git for version control and collaborative code development across distributed teams.

Practical notes

• This role is based in the San Jose office and requires on-site presence during standard business hours.
• Candidates should expect to work within a cross-functional team that includes data engineers, product managers, and risk analysts.
• The hiring process includes multiple rounds of technical interviews, including a live coding session and a case study presentation.
• BILL is an equal opportunity employer and welcomes applicants from all backgrounds, identities, and experience levels.