A large global financial institution is seeking a hands-on Data Modeler to support AI, machine-learning, and predictive-analytics initiatives within a cyber-focused organization.
Despite the title, this is not a traditional database-modeling position centered on ERDs or schema design. The role is focused on building, testing, and validating predictive and behavioral models using Python, PySpark, Databricks, and large-scale distributed datasets.
The successful candidate will contribute across the model lifecycle, including development, evaluation, troubleshooting, validation, documentation, and production-readiness assessment.
RESPONSIBILITIES- Build and refine predictive, statistical, behavioral, and machine-learning models.
- Develop modeling and testing logic using Python and PySpark.
- Work with Databricks and distributed data-processing technologies.
- Prepare, transform, and analyze model inputs and features.
- Design and execute model-performance tests.
- Independently validate model logic, assumptions, outputs, and behavior.
- Analyze false positives, false negatives, thresholds, and unexpected patterns.
- Investigate whether performance issues originate from data quality, features, transformations, methodology, or model logic.
- Adjust and retest models based on validation findings.
- Evaluate models against expected and unexpected behavioral patterns.
- Document modeling decisions, testing methodology, validation results, limitations, and recommendations.
- Explain technical methodology and findings to relevant stakeholders.
- Support models as they progress toward production readiness.
QUALIFICATIONS- Hands-on experience developing predictive, statistical, machine-learning, or behavioral models.
- Direct experience testing and validating models.
- Demonstrated ownership across multiple stages of the model lifecycle.
- Strong Python programming skills.
- Practical PySpark experience.
- Hands-on Databricks experience.
- Experience working with large, complex, or distributed datasets.
- Strong understanding of model evaluation and performance testing.
- Ability to identify and troubleshoot data-quality, feature-engineering, and model-performance issues.
- Strong quantitative and analytical problem-solving skills.
- Ability to document and explain modeling and validation decisions clearly.
- Ability to work onsite three days per week in Jersey City or Charlotte.
PREFERRED EXPERIENCE- Cybersecurity or insider-risk analytics.
- Fraud or anomaly detection.
- Behavioral or surveillance analytics.
- Banking or financial-services experience.
- Java.
- Model-drift analysis.
- Experience productionizing analytical models.
- Experience developing models within a regulated enterprise environment.