GPU Compute Cluster Engineer

Inferact; Singapore; On-site; SGD 200,000 - 400,000; Full time

Benefits offered by this job

Medical coverage

Dental coverage

Vision coverage

Equity

Job summary

Inferact is seeking a hands-on cluster administration engineer to own and operate high‑performance GPU compute infrastructure in Singapore. You will ensure health, availability, observability, and usability across neo-cloud and dedicated providers to accelerate vLLM development.

You will own GPU health, scheduling, diagnostics, and incident response, working with engineering leaders to standardize provisioning and operations. Strong Linux, HPC, and scripting skills are essential for success.

Qualifications

• BS degree or equivalent in CS/engineering or related field.
• Experience administering large compute clusters or HPC environments.
• Strong Linux admin fundamentals and debugging skills.
• Experience with GPU servers, drivers, and diagnostics.
• Familiarity with SLURM, Kubernetes, or similar tooling.
• Ability to own incidents end-to-end and automate ops workflows.

Responsibilities

• Own cluster health, GPU availability, monitoring, alerting, and scheduling.
• Operate across multi-provider GPU compute environments and memory errors.
• Standardize provisioning, operation, and debugging procedures.
• Collaborate with engineering leadership to scale compute for ML workloads.

Skills

Bachelor’s degree or equivalent in CS/

HPC clusters administration

Linux systems administration

GPU servers management

Cluster scheduling with SLURM

Kubernetes or equivalent tooling

Incident ownership and response

Automation with Bash/Python/Ansible

Education

Bachelor's degree or equivalent

Tools

SLURM

Kubernetes

Ansible

Terraform

Helm

Job description

Inferact is seeking a hands-on cluster administration engineer to own and operate high‑performance GPU compute infrastructure in Singapore. You will ensure health, availability, observability, and usability across neo-cloud and dedicated providers to accelerate vLLM development.

You will own GPU health, scheduling, diagnostics, and incident response, working with engineering leaders to standardize provisioning and operations. Strong Linux, HPC, and scripting skills are essential for success.