Fraud Analytics AI Engineer: Agentic ML Systems

FPT Asia Pacific Pte Ltd; Singapore; On-site; SGD 90,000 - 140,000; Full time

Job summary

FPT Asia Pacific Pte Ltd is seeking an ML/AI Engineer embedded in the data analytics and AI division to design and build agentic AI systems for fraud detection. You will deliver MVPs, integrate with Redshift/Glue/PySpark, and establish governance and monitoring across deployments.

The role requires Python expertise, experience with LLM-enabled solutions, and collaboration with data scientists, engineers and business stakeholders to operationalize fraud analytics capabilities.

Qualifications

• Degree in CS/AI/DS/SWE/IS or related discipline.
• 3-5 years hands-on ML/AI app development; at least one LLM-enabled system in production.

Responsibilities

• Design end-to-end fraud analytics agentic AI architecture.
• Develop AI agents to identify emerging fraud risks for officer review.
• Translate validated risks into actionable detection features and logic.
• Integrate agent workflows with fraud analytics platform data (Redshift, Glue, PySpark).
• Build modular, tested services with clear interfaces and version control.
• Create CI/CD workflows for testing, validation and deployment of AI components.
• Establish evaluation frameworks, governance controls and guardrails for AI systems.
• Implement observability and monitoring for system health, latency and data quality.
• Collaborate with data scientists, engineers and business stakeholders.
• Produce runbooks and technical documentation for long-term sustainability.

Skills

Python programming

LLM-enabled systems

Agentic AI

API development

Retrieval & RAG

Collaboration with stakeholders

Education

Degree in Computer Science, Artificial Intelligence, Data Science, Software Engineering, Information Systems or related

Tools

LangGraph

CrewAI

PySpark

Databricks

CI/CD tooling

Docker

MLflow

Job description

FPT Asia Pacific Pte Ltd is seeking an ML/AI Engineer embedded in the data analytics and AI division to design and build agentic AI systems for fraud detection. You will deliver MVPs, integrate with Redshift/Glue/PySpark, and establish governance and monitoring across deployments.

The role requires Python expertise, experience with LLM-enabled solutions, and collaboration with data scientists, engineers and business stakeholders to operationalize fraud analytics capabilities.