Global Payment Holding Company is building risk and fraud capabilities for a regulated payments environment, combining end-to-end ML delivery with AI-assisted, LLM/agentic workflows. This onsite role in Alpharetta, GA will help shape how detection models are designed, validated, deployed, monitored, and continuously optimized, while also creating practical analyst workflows that support fraud, credit, and AML operations.
The role focuses on turning regulatory requirements and operational needs into detection strategies and technical execution, partnering across Risk, Compliance, and Technology, and leading teams through complex, ambiguous detection challenges.
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
