Senior ML Scientist: Fraud Detection & Impact Analytics

Adobe; Washington (District of Columbia); Hybrid; USD 133,000 - 236,000; Full time

Job summary

Adobe is seeking a Senior Data Scientist to build fraud and abuse detection models and measure their effectiveness. You will develop robust evaluation frameworks, define ground truth, and communicate impact with executives.

The role blends hands‑on modeling with experimentation in a hybrid work setting requiring at least 3 days in the office. You should have 8+ years in applied Data Science/ML, strong Python/SQL, model monitoring, and a proven ability to translate complex analyses into

Qualifications

• 8+ years in applied Data Science / ML with production ML experience.
• Strong foundation in statistics and classical ML, experimentation, evaluation, and performance measurement.
• Strong Python and SQL skills with large, complex datasets.
• Experience with model monitoring, drift, false positives/negatives, and imperfect labels.
• Strong data visualization and storytelling to translate analysis into insights.
• Independent, owner‑minded with ability to navigate ambiguity.
• Bachelor's or equivalent; advanced degree preferred.

Responsibilities

• Build and tune ML models for fraud and abuse detection.
• Develop evaluation frameworks, datasets, and metrics for model effectiveness.
• Analyze false positives/negatives, drift, and emerging fraud patterns.
• Define ground truth, labeling approaches, and fraud taxonomies.
• Design experiments and evaluate tradeoffs across precision, recall, and impact.
• Build dashboards translating detection performance into business impact.
• Ensure data quality and guard against leakage and bias.
• Collaborate with engineering, product, policy, and risk teams.

Skills

Python

SQL

Data visualization

Storytelling

Statistics

Machine learning

Model monitoring

Experimentation

Education

Bachelor's degree in Statistics/Mathematics/CS

Advanced degree a plus

Job description

Adobe is seeking a Senior Data Scientist to build fraud and abuse detection models and measure their effectiveness. You will develop robust evaluation frameworks, define ground truth, and communicate impact with executives.

The role blends hands‑on modeling with experimentation in a hybrid work setting requiring at least 3 days in the office. You should have 8+ years in applied Data Science/ML, strong Python/SQL, model monitoring, and a proven ability to translate complex analyses into