Role OverviewWe are seeking a Risk Modeler to develop, validate, and operationalize the analytic models that quantify and prioritize risk within our unclassified PAI/CAI-based analytic platform. This role is responsible for designing statistical, machine-learning, and scoring models that turn multi-source data and graph relationships into defensible, explainable risk indicators for mission analysts and decision-makers. The ideal candidate pairs rigorous quantitative methods with a practical focus on explainability, uncertainty, and operational use.Key ResponsibilitiesModel DevelopmentDesign and develop statistical, probabilistic, and machine-learning models to quantify risk across entities, events, and networksBuild risk-scoring, ranking, and prioritization methodologies from multi-source dataDevelop anomaly-detection, forecasting, and pattern-analysis modelsIncorporate geospatial and temporal features into risk models where applicableValidation & ExplainabilityValidate models for accuracy, robustness, bias, and stability, including back-testing and sensitivity analysisEnsure models are explainable, defensible, and appropriately caveated for analytic useQuantify and communicate uncertainty, confidence, and the impact of data qualityDocument model assumptions, methodology, and limitationsOperationalizationWork with data and platform engineers to deploy models into production pipelinesMonitor model performance and drift; maintain retraining and evaluation workflowsTranslate analyst and mission requirements into clear model specificationsSecurity & ComplianceEnsure models and data handling meet security requirements for sensitive environmentsSupport Authority to Operate (ATO) processes, model governance, and compliance frameworksRequired QualificationsTechnical Expertise4+ years developing quantitative, statistical, or machine-learning modelsStrong programming skills in Python and its scientific stack (NumPy, pandas, scikit-learn); R a plusSolid foundation in statistics, probability, and quantitative methodsExperience with ML frameworks (scikit-learn, XGBoost, PyTorch, or TensorFlow)Experience with model validation, evaluation metrics, and back-testingAbility to work with large, messy, multi-source datasetsModeling BreadthExperience with risk scoring, anomaly detection, forecasting, or predictive modelingFamiliarity with explainability techniques (SHAP, LIME) and model documentationUnderstanding of uncertainty quantificationDomain KnowledgeExperience translating operational or analytic requirements into modelsAbility to communicate methodology and results clearly to non-technical stakeholdersPreferred QualificationsActive security clearance or ability to obtain oneExperience in government, defense, or intelligence contracting environmentsDomain experience in one or more of: supply chain risk, maritime or geospatial risk, or threat/security analyticsExperience with geospatial-temporal modeling (GIS, spatial statistics)Familiarity with graph-based features or network analytics as model inputsFamiliarity with PAI/CAI data sourcesAdvanced degree in a quantitative field (statistics, operations research, data science, applied mathematics, or economics)Technical EnvironmentLanguages: Python (R, SQL a plus)ML / Stats: scikit-learn, XGBoost, PyTorch/TensorFlow, statsmodelsExplainability: SHAP, LIME, model documentation practicesGeospatial: GIS and spatial-statistics tooling (where applicable)Infrastructure: Docker, Kubernetes, cloud platforms (AWS GovCloud, Azure Government), MLOps toolingSecurity: Secure data handling, model governance This role turns the platform's data and relationships into decision-ready signals: rigorous, transparent risk models that analysts and decision-makers can trust and defend.