Lead Risk Data Scientist & ML Engineer

Responsibilities

• Own the end-to-end delivery of detection models and AI-assisted workflows that support fraud, credit, and AML risk operations.
• Drive model performance across the full lifecycle, from design and validation to production deployment and continuous optimization.
• Translate regulatory requirements and operational needs into detection strategies and technical execution plans.
• Maintain model governance with explainable, auditable approaches aligned with regulatory expectations, including documentation and change management.
• Lead model development work including design, development, validation, deployment, and serving in production.
• Implement and oversee model performance monitoring using production data and investigation outcomes.
• Design scalable batch and real-time data pipelines that support model development and serving.
• Design and deploy AI-assisted analyst workflows using LLMs and agentic frameworks, and guide agent-based systems that augment human decision-making in risk operations.
• Work at the pilot/proof-of-concept stage to establish best practices for scaling detection and AI workflow solutions.
• Define and refine detection strategies based on emerging fraud patterns and regulatory requirements.
• Track and optimize key performance metrics including precision, recall, false positives, and alert quality.
• Make tradeoff decisions between detection coverage, operational cost, and false positive rates.
• Support compliance expectations (including BSA/AML, OFAC, FinCEN, SR 11-7) and partner with Model Risk Management and Compliance for validation and regulatory reviews.
• Serve as the primary technical partner to Fraud Operations, Compliance, and Technology teams.
• Lead cross-functional project teams from conception through deployment, establishing priorities, performance expectations, and delivery accountability.
• Provide technical guidance and mentorship to data scientists and engineers executing on risk initiatives.
• Build team capability across detection modeling, data engineering, and AI/agentic systems.

Requirements

• 7+ years in data science, machine learning, or MLOps.
• Proven experience developing, deploying, and maintaining detection models (fraud, AML, or credit risk) in production environments.
• Hands-on experience with AI-assisted workflows, LLMs, and agentic frameworks, including pilot-stage deployments.
• Experience in regulated financial services or fintech environments is preferred.
• Exposure to model risk management frameworks and regulatory interactions, including SR 11-7.
• Strong proficiency in Python and SQL.
• MLOps experience including Git, GitHub Actions, CI/CD practices, model monitoring, retraining pipelines, and infrastructure automation.
• Hands-on experience with data science platforms such as Databricks, Snowflake, and AWS SageMaker.
• AWS ecosystem expertise including SageMaker, Glue, Lambda, EventBridge, and related services.
• Familiarity with LLM and agentic frameworks including foundational models (Claude, GPT, etc.) and orchestration tools (AWS AgentCore, LangChain, etc.).
• Understanding of fraud typologies, AML transaction monitoring methodologies, and detection system design.
• Resourceful and versatile in a small, fast-moving team environment; comfortable delivering with constrained resources.
• Startup mindset with an emphasis on execution and measurable outcomes.
• Combines technical depth with collaborative leadership, guiding teams through ambiguous problems and driving clarity and delivery.
• Collaborates effectively across Risk, Compliance, and Technology, translating strategy into action.

Technologies

Python, SQL, Git, GitHub Actions, CI/CD, Databricks, Snowflake, AWS SageMaker, Glue, Lambda, EventBridge, Claude, GPT, AWS AgentCore, LangChain