The Financial Crimes Model Analyst plays a critical role in designing, validating, and optimizing analytical models that support financial crime prevention, including fraud, AML/KYC, and sanctions monitoring. This role leverages modern data-agnostic, low/no-code analytical platforms and machine-learning tooling to operationalize robust detection logic, integrate LLM/SLM outputs responsibly, and ensure alignment with Model Risk Management (MRM) standards.
PRINCIPAL DUTIES AND RESPONSIBILITIES
-
Designs, validates, and enhances financial crime detection models across fraud, AML/KYC screening, and related domains.
-
Applies statistical techniques to evaluate and calibrate LLM/SLM outputs when used in decision-support workflows.
-
Conducts ablation studies, back testing, and reproducible experiments to ensure model stability and business impact.
-
Optimizes model and pipeline efficiency, including latency, throughput, and computational performance.
-
Develops, tracks, and interprets performance metrics.
- Implements monitoring for drift, bias, degradation, and shifts.
-
Produces audit-ready MRM documentation, including validation plans, governance artifacts, explainability notes, and calibration reports.
-
Maintains model health dashboards and reporting for executives, governance bodies, and oversight functions.
-
Partners with internal and external data engineers and vendor partners to maintain reliable inputs and production workflows.
-
Develops SQL and Python assets for exploration, experimentation, reporting, and automated artifact generation.
-
Maintains clear and comprehensive documentation, including data dictionaries, model cards, decision logic, and workflow diagrams.
-
Works cross-functionally with engineering, product, compliance, risk, and vendor teams to deploy and maintain models.
-
Translates complex analytical concepts into accessible insights for varied technical and business audiences
-
Performs other duties as assigned.
EDUCATION AND/OR EXPERIENCE
-
Bachelor’s degree in Statistics, Econometrics, Data Science, Mathematics, Computer Science, or a related quantitative field, required; Master’s degree, preferred.
-
3-5 years’ experience in model development, evaluation, monitoring, risk governance, or model lifecycle management, required.
-
3-5 years’ experience in BSA/AML compliance, fraud and/or case investigation, or experience in quality assurance/control or internal audit, required.
-
Hands-on expertise in SQL and Python for analysis and rapid prototyping, required.
-
Familiarity with modern modeling techniques, or similar low/no-code platforms, and the responsible use of LLM/SLM capabilities, required.
-
Experience producing governance-grade validation or audit documentation, required.
-
Exposure to financial crime domains (AML, KYC, fraud), preferred.
-
CAMS and/or CAFP certifications, preferred.
KNOWLEDGE, SKILLS, AND ABILITIES
-
Strong understanding of supervised/unsupervised ML evaluation, calibration, drift detection, and explainability methods.
-
Knowledge of data quality domains and data lineage principles.
-
Familiarity with AI model governance concepts, including MRM standards and regulatory expectations.
-
Ability to develop and maintain dashboards and report for model health and risk indicators.
-
Knowledge of regulatory environment(s) and emerging BSA/AML and fraud trends.
-
Strong investigative, written, and oral communication skills.
-
Strong commitment to ethics, and the ability to understand a variety of issues and perspectives.
-
Understanding of the banking industry, including bank partnerships with fintech companies.
-
Multitasks effectively and takes action promptly, both independently and in a team environment.
-
Handles highly confidential information with appropriate discretion, and works well in a high volume, fast paced environment.
