Job Responsibilities:
- Deliver production AI/ML solutionsfor CCOR Conduct risk & compliance use cases by translating typologies, red flags, and control objectives into measurable model outcomes (e.g., precision/recall improvements, false-positive reduction, investigator efficiency).
- Drive & Execute research and applied innovation in supervised/unsupervised/semi‑supervised learning, graph/network analytics, anomaly detection, and weak supervision to improve true-positive rates, reduce false positives, and enhance investigator productivity.
- Develop and enhance detection modelsusing supervised/unsupervised/semi-supervised approaches (e.g., anomaly detection, clustering, weak supervision) and, where applicable,graph/network analyticsto identify complex patterns and relationships.
- Execute key parts of the model lifecycle: data sourcing (with appropriate controls), feature engineering (behavioral/temporal/entity/link features), model training, evaluation, calibration/thresholding, and performance monitoring.
- Implement interpretable ML and human-in-the-loop workflowsby supporting explainability (e.g., SHAP/LIME), stable reason codes, and feedback loops with investigators to improve usability and model precision over time.
- Contribute to MLOps and scalable deploymentby partnering with technology teams on CI/CD for ML, model registry usage, automated monitoring (data drift/concept drift), and repeatable, well-governed release processes.
- Support model risk management (MRM) deliverablesby producing documentation and analysis needed for validation (assumptions, limitations, benchmarking/challengers, back-testing, stability/drift analysis) and addressing review feedback.
- Collaborate across stakeholders(RCC, Investigations, Operations, Technology) to align on requirements, data readiness, controls, and target operating model for sustained production support.
- Apply GenAI/LLMs pragmatically(e.g., case narrative generation, unstructured text extraction/summarization) while prioritizing classical/statistical/graph ML methods where they deliver stronger, defensible detection efficacy.
Required qualifications, capabilities, and skills:
- Master's degree(or PhD preferred) in a quantitative discipline (Computer Science, Statistics, Mathematics, Economics, Operations Research, or related).
- Minimum 4 yearsof hands-on AI/ML experience, preferably with exposure tofinancial crime compliance / conduct risk / AML / fraud / sanctionsor similar control environments.
- Demonstrated experience building and/or deploying ML solutions (risk scoring, anomaly detection, triage/prioritization, NLP/LLM-enablement) with a focus on measurable outcomes.
- StrongPythonskills and experience with modern ML frameworks (e.g., PyTorch/TensorFlow) and common data/ML tooling.
- Practical knowledge of: imbalanced learning, cost-sensitive evaluation, feature engineering, model calibration/threshold optimization, and performance measurement in detection settings.
- Working knowledge ofMRM expectations(documentation, validation support, explainability, monitoring) in regulated financial services environments.
- Clear communication skills-able to explain model behavior, tradeoffs, and outputs (including reason codes) to technical and non-technical stakeholders.
- Ability tomentor junior team membersthrough code reviews, pairing, and technical guidance.
