Remote Fraud Data Scientist — Build production ML models

Experian; United States; Remote; USD 104,000 - 180,000; Full time

Benefits offered by this job

Bonus plan

Medical/dental/vision coverage

Flexible work options (remote/hybrid)

Paid holidays & time off

Job summary

Experian is seeking a Data Scientist to build fraud detection models and features identifying high-risk activity while minimizing friction for legitimate customers.

You'll collaborate with senior data scientists and engineers from problem definition through feature engineering, experimentation, and deployment, translating theory into production-ready solutions in a distributed data environment.

Qualifications

• At least 3 years of data science or ML experience.
• BS/MS in a quantitative field.
• Direct experience with fraud-detection or adversarial risk models.
• Experience creating meaningful fraud features.
• Proficiency in Python and PySpark; scalable code for large datasets.

Responsibilities

• Investigate large datasets to identify fraud patterns and signals.
• Translate fraud problems into hypotheses, plans, and metrics.
• Develop ML models for fraud detection across accounts and identity risk.
• Evaluate models using ROC/AUC, KS, Gini, and precision/recall metrics.
• Develop predictive features using identity, transactional, and network data.
• Write clean, tested Python and PySpark code for production environments.
• Monitor model performance and drift; adapt as needed.
• Present analyses on model behavior, risks, and recommendations.
• Follow data privacy, explainability, and governance standards.

Skills

Python

PySpark

Pandas

NumPy

scikit-learn

XGBoost

TensorFlow

Feature engineering

Model evaluation

Fraud analytics

Education

Bachelor's or higher in CS/Statistics/Engineering/Data Science

Tools

PySpark

Job description

Experian is seeking a Data Scientist to build fraud detection models and features identifying high-risk activity while minimizing friction for legitimate customers.

You'll collaborate with senior data scientists and engineers from problem definition through feature engineering, experimentation, and deployment, translating theory into production-ready solutions in a distributed data environment.