Senior Data Scientist - Fraud Detection

DataVisor•Mountain View, CA

8h•$140,000 - $170,000

About The Position

DataVisor is seeking a senior-level Data Scientist to join its Fraud Detection team. This dual-track role combines developing production machine learning systems for large-scale fraud detection with personally investigating specific fraud attacks. The role involves reconstructing attacker behavior and turning case-level findings into trend reports and improvements to fraud detection. The Data Scientist will lead the lifecycle of fraud detection features and models, including ideation, data exploration, prototyping, productionization, and monitoring. They will engineer predictive features from complex data such as user behavior, device intelligence, network graphs, and transaction records, and work with massive, noisy, imbalanced datasets using tools such as Spark, SQL, and DataVisor's proprietary AI platform. The role also uses agentic AI to automate analytics and create reusable tools for investigation, feature generation, and reporting. The position includes leading complex fraud investigations across identities, accounts, devices, and transactions; reconstructing attacker sequences and assessing intent and tooling; and producing evidence-backed technical reports, case studies, and customer-facing fraud trend reports for a range of stakeholders.

Requirements

Master's or PhD in Computer Science, Statistics, Mathematics, or a related quantitative field.
At least 3 years of applied experience in fraud detection, cybersecurity, or a related adversarial or high-velocity risk domain, such as fintech, consumer payments, banking, SaaS, marketplace risk, or security research.
Solid understanding of classic machine learning models, such as Logistic Regression and Gradient Boosting.
Hands-on experience with the machine learning lifecycle in a production environment.
Demonstrated investigator mindset, including pattern synthesis, hypothesis testing, and triaging signal from noise in ambiguous, adversarial cases.
Strong Python programming skills and proficiency with SQL.
Experience with large-scale data tools such as Spark or Hadoop and cloud platforms such as AWS, GCP, or Azure.
Excellent communication skills, including the ability to explain complex, ambiguous, or technical behavior to technical and non-technical audiences, customers, and executives.
Professional proficiency in written and spoken English and ability to collaborate effectively in a global, cross-functional team.

Nice To Haves

Experience with PySpark.

Responsibilities

Lead the full lifecycle of fraud detection features and models, from ideation and data exploration through prototyping, productionization, and monitoring.
Develop predictive features from large-scale, multidimensional data, including user behavior, device intelligence, network graphs, and transaction records.
Process massive, noisy, and imbalanced datasets using tools such as Spark, SQL, and DataVisor's proprietary AI platform.
Use agentic AI to automate analytic pipelines and develop reusable tools for fraud investigation, feature generation, and reporting.
Lead investigations into complex fraud cases spanning identities, accounts, devices, and transaction surfaces.
Reconstruct attacker sequences and develop hypotheses about actor intent and tooling.
Produce clear, evidence-backed technical reports and case studies for product, engineering, operations, legal, and executive stakeholders.
Generate customer fraud trend reports that synthesize case-level findings and aggregate data into actionable narratives.