E
Experian
Experian colaborează cu branduri recunoscute în domeniile experienței digitale, tehnologiei și transformării afacerilor. Activitatea sa combină informații despre clienți, strategia de produs, soluții tehnologice agile, inteligență artificială, arhitectură cloud și management de proiect, pentru a ajuta organizațiile să îmbunătățească experiențele clienților și să își eficientizeze operațiunile. Orientarea amplă a companiei către tehnologie și consultanță poate fi atractivă pentru profesioniștii interesați de roluri care acoperă transformarea digitală, marketingul, logistica și servicii de business conexe.

Expert Data Scientist, Fraud Risk Detection at Experian (Remote, US)

Build and assess machine learning models and predictive features to detect fraud and identity risk at Experian. The role uses Python and PySpark and supports deployment in batch, retrospective, and real-time decisioning systems.

Descriere

  • Analyze large datasets and develop fraud labels through exploratory analysis
  • Identify fraud patterns, attack techniques, and behavioral signals
  • Turn unclear fraud and risk challenges into hypotheses, analysis plans, model specifications, and measurable success criteria
  • Build machine learning models to detect account-opening fraud, account takeover, and identity risk
  • Assess models using ROC/AUC, KS, Gini, precision and recall, fraud capture, false positives, customer friction, and prevented losses
  • Create and validate predictive features from identity, transaction, credit history, device, behavioral, temporal, velocity, network, and third-party data
  • Write efficient, well-tested Python and PySpark code
  • Work with teams to deploy models and features in batch, retrospective, or real-time decisioning systems
  • Track feature quality, model performance, population shifts, and changes in fraud patterns
  • Present analyses of model behavior, tradeoffs, risks, and recommendations
  • Apply data privacy, documentation, explainability, validation, and governance standards
  • Report to the Senior Manager of Fraud Analytics
  • Partner with senior data scientists and engineers from problem definition through feature engineering, experimentation, and deployment

Cerințe

  • At least three years of experience in data science, machine learning, statistical modeling, or a related quantitative field
  • Bachelor’s or advanced degree in computer science, statistics, engineering, data science, or another quantitative discipline
  • Hands-on experience developing models for fraud detection, identity risk, credit risk, financial crime, or other adversarial risks
  • Experience creating useful fraud-detection features
  • Proficiency in Python and PySpark
  • Experience writing modular, tested code for large datasets and distributed or cloud data systems
  • Experience with pandas, NumPy, scikit-learn, XGBoost, TensorFlow, or similar tools
  • Knowledge of supervised learning, model evaluation, feature selection, statistical inference, experimentation, and model calibration
  • Experience handling class imbalance, delayed or incomplete labels, changing attack patterns, and model drift
  • Experience deploying models to production beyond the experimentation stage
  • Meet applicable United States work authorization and employment requirements

Beneficii

  • Competitive compensation package and bonus plan
  • Medical, dental, and vision coverage, plus a matching 401(k)
  • Flexible work options, including remote, hybrid, or office-based work
  • Flexible time off, including volunteer time, vacation, sick leave, and 12 paid holidays
  • Comprehensive benefits package
  • Opportunity for variable pay
  • Inclusive, purpose-driven culture
  • Accommodation support for disabilities or special needs

Locuri de muncă similare