Senior / Principal MLOps Engineer

About PayNet:

Payments Network Malaysia (PayNet) is Malaysia's premier payments network and central infrastructure for financial markets. We innovate, build and operate world-class payment systems and financial market infrastructures that safely, reliably and efficiently enable the functioning and development of Malaysia's financial system as well as the economy as a whole.

About the Risk & Compliance Division:

Managing risk to enable innovation without compromising trust. RCD identifies, manages, and mitigates emerging risks, safeguarding compliance and reinforcing organizational resilience across the payment ecosystem.

Summary of responsibility:

Responsible for designing, operating, and continuously enhancing the MLOps and cloud infrastructure that powers PayNet’s national-level fraud intelligence platforms and related services. This role ensures that models, data pipelines, and scoring services are secure, scalable, explainable, and production-grade, and supports controlled deployments, operational stability, and ongoing model lifecycle management in collaboration with key stakeholders.

Key responsibilities:

1. Own end-to-end MLOps productionisation for fraud and risk intelligence use cases, including packaging models into deployable APIs or batch services and enabling reliable production operations.
2. Build and maintain CI/CD pipelines for model and service build, testing, deployment, and controlled release/rollback across non-production and production environments.
3. Implement model lifecycle controls including model versioning, artifact management, experiment tracking, and support retraining/recalibration and controlled redeployment cycles in collaboration with Data Scientists.
4. Architect and operate secure cloud and hybrid infrastructure (AWS and on-premises) to support fraud data and machine learning platforms, ensuring scalability, resilience, and high availability.
5. Implement platform engineering best practices including containerisation, Infrastructure-as-Code, environment segregation, secrets management, and IAM governance with strict access controls.
6. Ensure auditability, security, and compliance of deployments and infrastructure changes, meeting regulatory and IT security requirements while optimising runtime stability, performance, and resource efficiency.

What will make you successful?

Must-have:

1. 5–8 years of hands-on experience in MLOps, DevOps, Cloud Engineering, or production ML systems, with demonstrated ability to deploy and operate production-grade ML/scoring services and pipelines.
2. Proven capability to design and operate secure, scalable, and resilient platforms in AWS and/or hybrid environments, including strong fundamentals in deployment automation and controlled environment promotion.

Advantage to have:

1. Experience with AWS cloud environments, including Terraform (IaC), Kubernetes/container orchestration, and Helm-based deployment management across Dev/UAT/Production.
2. Experience designing and maintaining CI/CD pipelines using GitLab, including infrastructure workflow controls (e.g., Atlantis-style gated changes).
3. Familiarity with distributed computing (e.g., Spark, Ray) and workflow orchestration tools (e.g., Prefect, Airflow) for scalable ML/batch workloads.
4. Experience operating or supporting JupyterHub or similar collaborative ML development environments, and transitioning work from experimentation to production pipelines.