Data Scientist - Career

Equifax•Alpharetta, GA

About The Position

As a Data Scientist (P3) on the Fraud Modeling Team, you will lead the design, development, and deployment of high-impact machine learning and advanced statistical models to combat First-Party Fraud, Third-Party Fraud, Synthetic Identity Fraud, and support Identity Graph Analytics. You will operate independently in a collaborative, matrixed environment, bridging technical model execution with commercial outcomes. You will work with Product, Sales, Core Analytics, and Model Risk Management (MRM) to build stress-tested, deployable fraud detection engines, support regulatory alignment, and communicate technical business impact to leadership. The role combines programming, data management, and statistical analysis to develop prototypes and derive insights from large and varied datasets. It also includes querying and structuring large historical datasets, engineering behavioral features and fraud signals, maintaining modular code and version-control practices, contributing to internal repositories, reviewing peers’ code, presenting complex analyses to technical and non-technical audiences, and mentoring junior data scientists.

Requirements

  • Master’s degree or equivalent practical experience in a highly quantitative STEM field such as Statistics, Computer Science, Physics, Mathematics, Engineering, or Data Science.
  • Hands-on experience with graph neural networks or graph databases such as Neo4j or NetworkX for entity resolution and graph-based feature engineering.
  • 5 to 7 years of hands-on data science experience, with direct focus on predictive models for fraud risk, identity resolution, or risk management.
  • Deep expertise in tree-based ensembles such as XGBoost and LightGBM, and modern deep learning models such as Transformers, neural networks, and sequence models.
  • Advanced Python skills, including Pandas, NumPy, Scikit-Learn, and PyTorch or TensorFlow.
  • Expert-level SQL query development for complex data extraction and feature engineering.
  • Proficiency with version control software, CI/CD practices, modular code design, and statistical repositories.
  • Strong grounding in model stress testing, cross-validation for highly imbalanced target distributions, model explainability, and performance tuning.
  • Ability to communicate complex machine-learning concepts clearly to non-technical stakeholders and align technically with Model Risk Management.
  • Ability to deliver moderately complex, production-ready modeling pipelines with speed, accuracy, and minimum operational supervision.

Nice To Haves

  • Experience with cloud deployment workflows in AWS, GCP, or Azure Data Lake / SageMaker.
  • Cloud certification preferred.
  • Familiarity with Model Risk Management documentation standards, such as SR 11-7 guidelines, and model governance protocols.

Responsibilities

  • Independently design, train, and optimize predictive models using XGBoost, Gradient Boosted Trees, Transformer-based models, and deep learning architectures for high-dimensional, imbalanced fraud datasets.
  • Use graph structures and identity link analysis to detect synthetic identity networks, fraud rings, and suspicious identity resolution patterns across large datasets.
  • Develop actionable fraud detection frameworks for First-Party Fraud, Third-Party Fraud, and Synthetic Identity Fraud.
  • Partner with Product and Sales to translate market requirements into technical model specifications and communicate technical requirements back to business partners.
  • Collaborate with Model Risk Management to support model validation, conceptual soundness checks, and compliance reviews.
  • Query, manipulate, and structure large historical datasets using SQL and Python to engineer behavioral features and fraud signals.
  • Write clean, modular, efficient code; follow version-control best practices; contribute to internal code repositories; and conduct code reviews.
  • Present statistical results through visual narratives and executive presentations for leadership and non-technical stakeholders.
  • Mentor junior data scientists on modeling standards.
  • Maintain compliance and ownership of model performance and data integrity, and deliver production-ready modeling pipelines on time.

Benefits

  • Comprehensive compensation and healthcare packages
  • 401k matching
  • Paid time off
  • Organizational growth potential through an online learning platform with guided career tracks
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service