AI Cloud Platform Reliability Engineer

FIRMUS METAL INTERNATIONAL PTE. LTD.; Singapore; On-site; SGD 100,000 - 170,000; Full time

Job summary

Firmus runs large-scale, state-of-the-art AI infrastructure and operates the platform services underpinning the Firmus AI Cloud in production. The Platform Reliability Engineer will handle incidents across Linux, Kubernetes, storage, and virtualization, owning responses from detection to resolution with a focus on automated remediation and runbooks.

The role involves operating the multi-tenant control plane, virtualisation platform, and high-performance storage, ensuring service levels and rapid

Qualifications

• Solid experience operating production infrastructure in a 24/7 environment using site reliability engineering practices.
• Strong working knowledge of Linux systems administration and troubleshooting.
• Working experience with scale-out, parallel or enterprise storage.
• Familiarity with a virtualisation platform (e.g., Proxmox, VMware or KVM).
• Working experience using observability tools.

Responsibilities

• Triage, diagnose and resolve incidents across the stack, owning each incident from escalation to resolution.
• Operate the multi-tenant control and management plane that supports separation between tenant workloads and data.
• Operate the virtualisation platform and shared services to their declared service levels.
• Operate exabyte-scale distributed storage and S3-compatible object storage to service levels, escalating as needed.
• Manage the GPU compute fleet in production and support provisioning and validation activities.
• Contribute to automation by turning fixes into runbooks and automated remediation.
• Diagnose performance issues across the datapath and coordinate with network/storage specialists.
• Deliver changes as code with review, testing, rollout and rollback plans.
• Use the observability stack to track SLIs, query metrics, logs and traces, and improve dashboards.
• Execute patching and vulnerability remediation across services.
• Manage vendor support cases and drive to resolution with evidence.
• Participate in technical recovery during major incidents and blameless post-incident reviews.

Skills

Site reliability engineering

Linux systems administration

Scale-out storage

Virtualisation

Observability

Kubernetes

Tools

Proxmox

VMware

KVM

Ceph

Lustre

GPFS

NetApp

Job description

Firmus runs large-scale, state-of-the-art AI infrastructure and operates the platform services underpinning the Firmus AI Cloud in production. The Platform Reliability Engineer will handle incidents across Linux, Kubernetes, storage, and virtualization, owning responses from detection to resolution with a focus on automated remediation and runbooks.

The role involves operating the multi-tenant control plane, virtualisation platform, and high-performance storage, ensuring service levels and rapid