AI DevOps Engineer / MLOps Engineer

About the role

Our client, a leading Singapore‑based financial institution, is driving an enterprise‑wide AI transformation. They need an experienced AI DevOps/MLOps Engineer to design, automate, and operate model deployment pipelines that deliver secure, scalable AI solutions across the bank.

Key responsibilities

• Design, build and maintain CI/CD pipelines for AI/ML model deployment.
• Deploy Large Language Models from Hugging Face and other repositories into dev, UAT and production environments.
• Automate model packaging, versioning, testing and release using DevOps best practices.
• Manage Kubernetes/OpenShift AI platforms for model hosting and inference.
• Containerise AI applications with Docker and deploy to Kubernetes clusters.
• Integrate AI solutions with enterprise applications via REST APIs and micro‑services.
• Implement Infrastructure as Code (Terraform/Ansible) for AI platform provisioning.
• Monitor model health and application performance using Prometheus, Grafana, ELK, Dynatrace or Splunk.
• Optimise GPU utilisation, troubleshoot performance issues and ensure compliance with banking security standards.

Required profile

• 6–10 years of experience in DevOps, Platform Engineering or Cloud Engineering.
• Strong hands‑on experience with Jenkins and modern CI/CD tools (GitHub Actions, GitLab CI, Bitbucket).
• Proven experience deploying AI/ML models from Hugging Face or similar repositories.
• Deep knowledge of Docker, Kubernetes/OpenShift and Linux system administration.
• Solid scripting skills in Bash, Python or Shell.
• Experience with Terraform or Ansible for infrastructure automation.
• Familiarity with cloud platforms (AWS, Azure, GCP) and GPU‑enabled environments.

Required skills

• Jenkins
• GitHub, GitHub Actions, GitLab, Bitbucket
• Hugging Face model repositories
• Docker
• Kubernetes / OpenShift
• Linux
• Bash, Python, Shell scripting
• Terraform, Ansible
• Artifactory, Nexus, Docker Registry
• AWS, Azure, GCP
• NVIDIA CUDA / GPU environments
• Prometheus, Grafana, ELK, Dynatrace, Splunk