Operational Risk Modeler: Explainable ML & Scoring

3GIMBALS; Quantico (VA); On-site; USD 135,000 - 175,000; Full time

Job summary

3GIMBALS is seeking a Risk Modeler to design, validate, and operationalize analytic models that quantify risk across entities, events, and networks.

The role focuses on explainable, uncertainty-aware models using Python and ML frameworks, with production deployment and ongoing monitoring in a security-conscious environment. This position requires strong statistical foundations and the ability to translate analytic needs into robust model specifications.

Qualifications

• 4+ years developing quantitative, statistical, or ML models.
• Strong Python and scientific stack (NumPy, pandas, scikit-learn); R a plus.
• Solid statistics, probability, and quantitative methods.
• Experience with ML frameworks (scikit-learn, XGBoost, PyTorch, or TensorFlow).
• Experience with model validation, evaluation metrics, and back-testing.
• Able to work with large, messy, multi-source datasets.

Responsibilities

• Design and develop statistical, probabilistic, and ML models to quantify risk across entities, events, and networks.
• Build risk-scoring, ranking, and prioritization methodologies from multi-source data.
• Develop anomaly-detection, forecasting, and pattern-analysis models.
• Incorporate geospatial and temporal features into risk models where applicable.
• Validate models for accuracy, robustness, bias, and stability; ensure models are explainable.
• Document model assumptions, methodology, and limitations.

Skills

Quantitative modeling

Python programming

Statistics & probability

ML frameworks

Data handling

Model validation

Education

Advanced degree in quantitative field

Tools

NumPy

pandas

scikit-learn

XGBoost

PyTorch

TensorFlow

R

statsmodels

GIS

Docker

Kubernetes

AWS GovCloud

Job description

3GIMBALS is seeking a Risk Modeler to design, validate, and operationalize analytic models that quantify risk across entities, events, and networks.

The role focuses on explainable, uncertainty-aware models using Python and ML frameworks, with production deployment and ongoing monitoring in a security-conscious environment. This position requires strong statistical foundations and the ability to translate analytic needs into robust model specifications.