Senior Quantitative Developer Machine Learning & Regulatory Credit Risk

Key Responsibilities:
Develop and implement regulatory credit risk models (PD, LGD, EAD) using Python, Spark
(Scala), and distributed systems in a Kubernetes-based Azure environment.
Build scalable ML pipelines integrated with MLflow, CI/CD (Azure DevOps), and model
governance frameworks.
Create model explainability layers using tools such as SHAP, LIME, or custom
counterfactual frameworks to support model governance and audit.
Participate in the lifecycle of CECL and CCAR models, including data preparation, feature
engineering, model development, and documentation for Model Risk Governance
(MRG).
Partner with data engineers and risk modeling teams to ingest, process, and version
complex credit datasets from enterprise systems.
Conduct model validation, robustness testing, scenario analysis, and performance
monitoring in compliance with SR 11-7, OCC, and Fed requirements.
Lead efforts to incorporate alternative and unstructured data sources, including text
analytics and ESG data, into existing model frameworks.
Required Skills & Experience:
10+ years in quantitative development or model risk analytics, preferably in banking,
regulatory modeling, or enterprise risk domains.
Advanced expertise in:
o Python (NumPy, pandas, scikit-learn, PyTorch/TensorFlow)
o Apache Spark (Scala) for distributed ML workloads
o Azure Kubernetes Services (AKS), Terraform, MLflow
Deep understanding of U.S. regulatory frameworks: Basel III/IV, CECL, SR 11-7, SR 15-
18/19, and CCAR.
Proven experience building interpretable ML models and documenting them for use in
audited and regulated environments.
Strong communication skills for cross-functional collaboration with MRG, internal audit,
compliance, and technology teams.
Degree in a quantitative discipline such as Mathematics, Computer Science, Financial
Engineering, or Statistics (PhD or Master s preferred).
Prior work with regulatory capital model development or validation teams.
Familiarity with risk modeling architecture, tools, or data pipelines (Athena, Quartz).
Experience implementing AI/ML model fairness, bias detection, and transparency
controls in regulated environments.
Participation in regulatory exams (OCC, Federal Reserve, FDIC) or model submission
cycles.
Background in text mining, survival modeling, or NLP for financial documents is a plus.