AI Infra DevOps Engineer - GPU/Cloud HPC

The Supreme HR Advisory Pte Ltd; Singapore; On-site; SGD 57,000 - 77,000; Full time

Job summary

The Supreme HR Advisory Pte Ltd is seeking an AI Infrastructure Engineer to architect and operate high-density GPU compute clusters in a Singapore on-site setting. You will deploy and manage containerized AI/ML workloads using Kubernetes, Slurm, or Ray, while tuning Linux systems and GPUs for peak performance.

You will build automated pipelines with Terraform and Ansible, monitor system health with Prometheus/Grafana, and collaborate with AI/ML teams to minimize bottlenecks during distributed

Qualifications

• Proficient Linux systems administration, kernel tuning, and scripting (Bash/Python).
• Hands-on GPU HPC knowledge with CUDA and NVLink technologies.
• Experience with GPU container orchestration (Kubernetes) and HPC schedulers (Slurm, Run:ai, Ray).
• Familiar with high-speed networking and RDMA (RoCE v2/InfiniBand).
• Proficient in Terraform and Ansible for IaC and configuration management.

Responsibilities

• Architect, configure, and maintain high-density GPU compute clusters.
• Design and optimize low-latency networks and scalable storage.
• Build automated deployment pipelines using Terraform, Ansible, and Helm.
• Monitor performance, diagnose bottlenecks, and lead incident response.
• Collaborate with AI/ML teams on distributed training workloads.

Skills

Linux administration

Shell scripting

GPU/ HPC

Kubernetes

Networking (RDMA)

Terraform/Ansible

Education

Bachelor's degree in Computer Science/IT/Engineering

Tools

Kubernetes

Slurm

Ray

Terraform

Ansible

Ceph

Job description

The Supreme HR Advisory Pte Ltd is seeking an AI Infrastructure Engineer to architect and operate high-density GPU compute clusters in a Singapore on-site setting. You will deploy and manage containerized AI/ML workloads using Kubernetes, Slurm, or Ray, while tuning Linux systems and GPUs for peak performance.

You will build automated pipelines with Terraform and Ansible, monitor system health with Prometheus/Grafana, and collaborate with AI/ML teams to minimize bottlenecks during distributed