Senior GPU HPC Infra Engineer | Kubernetes & Slurm

MTAI Sdn. Bhd.; Kuala Lumpur; On-site; MYR 180,000 - 300,000; Full time

Job summary

MTAI Sdn. Bhd. seeks an experienced Infrastructure/Datacenter Engineer to own remote management of NVIDIA GPU servers and HPC fabric. You will coordinate hardware vendors, manage lifecycle processes and help scale the platform with HPC interconnects and orchestration tooling.

Ideal candidates have 8+ years in a similar role, hands-on GPU HPC ops, and strong Linux expertise. The role involves startup-like ownership in a small, fast-moving team in Kuala Lumpur.

Qualifications

• 8+ years in infrastructure, systems or datacenter engineering.
• Hands-on remote datacenter server management experience.
• Experience managing hardware/OEM vendor relationships.
• Fluency in HPC environments and InfiniBand interconnects.

Responsibilities

• Own remote management, monitoring and troubleshooting of datacenter servers, primarily NVIDIA GPU servers.
• Manage datacenter, hardware and OEM vendor relationships including procurement and support escalation.
• Support HPC infrastructure, including high-speed interconnects (InfiniBand) at fabric level.
• Deploy, configure and operate Kubernetes and Slurm for GPU workloads.
• Define standards and processes for hardware provisioning and server lifecycle management.
• Diagnose hardware, firmware and low-level infra issues with vendor coordination.
• Collaborate with Head of Sovereign AI Infra and infra team to ensure integrated stack.
• Contribute to capacity planning and scaling as platform grows.

Skills

Datacenter management

Vendor management

Kubernetes

Slurm

GPU compute infra

Linux

InfiniBand HPC networking

Firmware management

Capacity planning

Automation

Job description

MTAI Sdn. Bhd. seeks an experienced Infrastructure/Datacenter Engineer to own remote management of NVIDIA GPU servers and HPC fabric. You will coordinate hardware vendors, manage lifecycle processes and help scale the platform with HPC interconnects and orchestration tooling.

Ideal candidates have 8+ years in a similar role, hands-on GPU HPC ops, and strong Linux expertise. The role involves startup-like ownership in a small, fast-moving team in Kuala Lumpur.