The organization is focused on verifying good identities in real time and eliminating identity fraud online
Requirements
- Experience working with fraud or fraud-related datasets
- Proficiency in Python (preferred) or R, with experience in machine learning libraries such as scikit-learn, TensorFlow, PyTorch, or XGBoost
- Ability to analyze, clean, and model large-scale datasets using SQL and tools like AWS, Databricks, Hadoop, or Spark
- Experience creating dashboards in AWS Quicksight and Databricks
- Knowledge of supervised and unsupervised learning, feature engineering, and model evaluation
- Experience translating business challenges into data science solutions and effectively communicating results
Responsibilities
- Design and implement machine learning models and statistical algorithms for first party fraud and identity verification using diverse large-scale data sources
- Analyze large datasets to identify fraud patterns and opportunities for product improvements
- Utilize feedback, outcome, and fraud contribution data to enhance models and products
- Develop data processing pipelines, automated workflows, and tools for data cleansing, integration, and evaluation
- Provide analytical support to the fraud and risk data science team with clear communication of insights to technical and non-technical audiences
- Continuously test and apply new machine learning algorithms and techniques to improve models
- Build, maintain, and monitor scalable models in production; participate in code reviews and peer discussions
Other
- Master’s degree or higher in Computer Science, Mathematics, Statistics, or related quantitative field, or equivalent professional experience
- Collaborate with product, engineering, and cross-functional teams to develop data-driven solutions aligned with business objectives
- Contribute to a collaborative team environment by identifying trends and anomalies that inform broader product strategies
- Effectively communicate results to technical and non-technical audiences
- Participate in code reviews and peer discussions
