Senior ML Scientist: Fraud Detection & Impact

Adobe; New York (NY); Hybrid; USD 163,000 - 236,000; Full time

Benefits offered by this job

Hybrid work model

Excellent benefits

Job summary

Adobe is seeking a Senior Data Scientist, ML—Fraud Detection & Effectiveness in New York to build fraud and abuse detection models and quantify their business impact. You will own end-to-end modeling—from ground-truth definition and feature engineering to evaluation, monitoring, and executive-ready dashboards that explain results.

This is a hybrid role with minimum three days in the office. The role requires 8+ years in applied DS/ML, strong Python and SQL, and the ability to turn complex

Qualifications

• 8+ years in applied Data Science/ML with production models.
• Strong Python and SQL skills with large datasets.
• Experience with model evaluation, drift, and monitoring.
• Ability to translate analysis into actionable business insights.

Responsibilities

• Build and tune ML models for fraud and abuse detection.
• Develop evaluation frameworks and metrics for performance measurement.
• Analyze false positives/negatives, model drift, and emerging patterns.
• Define ground-truth labeling approaches and fraud taxonomies.
• Design experiments and evaluate tradeoffs across precision, recall and business impact.
• Build dashboards translating detection performance into business insights.
• Identify data quality issues and leakage sources.
• Collaborate with engineering, product, policy, and risk teams to implement improvements.

Skills

Python

SQL

Statistics

Experimentation

Data visualization

Communication

Education

Bachelor's in Statistics/Math/CS or related field

Advanced degree (MS/PhD) preferred

Job description

Adobe is seeking a Senior Data Scientist, ML—Fraud Detection & Effectiveness in New York to build fraud and abuse detection models and quantify their business impact. You will own end-to-end modeling—from ground-truth definition and feature engineering to evaluation, monitoring, and executive-ready dashboards that explain results.

This is a hybrid role with minimum three days in the office. The role requires 8+ years in applied DS/ML, strong Python and SQL, and the ability to turn complex