N

Nscale

Develops and operates AI data centers

AI Reliability Engineer 1

Site Reliability EngineeringUpdated on Nov 17
Full-timeSingaporeNo salary listed

About the Job

Requirements

  • At least 3 years of infrastructure support or support engineering experience, including support or service desk experience in structured, service-level-agreement-driven, customer-facing environments such as cloud, data centre, or managed services.
  • Working knowledge of GPU infrastructure, including hands-on use of nvidia-smi or similar diagnostics, interpreting hardware error output and logs, physically troubleshooting servers through component reseating and swap testing, working via baseboard management controller or out-of-band management, and preparing return merchandise authorization evidence.
  • Confident command-line interface skills with systemd, filesystems, permissions, and standard networking tools, with the ability to troubleshoot common issues independently and identify when to escalate.
  • Knowledge of IP addressing, subnets, virtual local area networks, routing, domain name system, and firewalls.
  • Experience with structured support processes such as ITIL, including prioritisation, escalation, service-level-agreement awareness, and accurate documentation.
  • Ability to use dashboards and alerts to identify symptoms, gather evidence, and follow runbooks, and to propose simple alert or dashboard improvements with review.
  • Ability to read and write simple Bash or Python scripts and use Git for version control.
  • Understanding of servers, networks, storage, and virtualisation concepts, ideally from a support or operations background.
  • Ability to participate in on-call duties after onboarding and travel when needed.

Responsibilities

  • Join the Support duty rotation and handle day-to-day tickets and alerts, escalating early and appropriately and collaborating with Engineering, with guidance, when incidents or changes require it.
  • Perform GPU node triage and hardware troubleshooting by interpreting nvidia-smi or Data Center GPU Manager output and system logs; isolating faults across GPU, network interface card, and server hardware; carrying out physical remediation such as reseats, swap testing, and component checks; and preparing evidence for vendor return merchandise authorisation.
  • Run fabric and link diagnostics following established runbooks, using mlxlink or an equivalent, capture evidence accurately, and provide handovers that allow the next engineer to continue the work.
  • Assist with storage and data-path investigations, including mounts, connectivity, and client-side symptoms, on high-performance platforms, gathering evidence for senior- or engineering-led diagnosis.
  • Follow established runbooks to resolve common issues, propose improvements, and contribute incremental fixes with review.
  • Record, update, manage, and resolve tickets, keeping all parties informed through clear notes, next steps, and customer communications via agreed channels.
  • Participate in monitoring, troubleshooting, and triage, capturing logs and facts to enable efficient handover.
  • Participate in changes under peer review, learning risk assessment and backout practices in live customer environments.
  • Maintain source-of-truth accuracy across data centre infrastructure management, inventory, and asset records using NetBox or similar patterns.
  • Identify automation opportunities and contribute simple scripts and tooling improvements to optimise processes.
  • Act as the escalation point for onsite data centre operations staff and coordinate smart-hands tasks within scope.
  • Learn platform fundamentals to help customers get value from services, seeking support when deeper expertise is needed.
  • Document validated steps, contribute to training materials, and shadow senior colleagues during complex work to build capability.
  • Contribute to incident reviews and help track preventative follow-ups within scope.
  • Deliver assigned tasks and project work to agreed quality and timelines, flagging blockers early and seeking help when needed.
  • Participate in scheduled on-call and out-of-hours work after onboarding; travel to Nscale or customer locations for deployments, troubleshooting, and operational tasks; and attend supplier training as required.

Desired Qualifications

  • Exposure to Remote Direct Memory Access or InfiniBand, link-level diagnostics such as mlxlink or ibdiagnet, NCCL-based troubleshooting, or NVLink concepts.
  • Exposure to VAST or comparable AI-optimised storage platforms, Ceph, or Network File System at scale, including basic storage-network troubleshooting.
  • Familiarity with OpenStack troubleshooting flows or fleet-scale provisioning and health tooling such as MAAS, NetBox, or Redfish.
  • Understanding of core container orchestration concepts such as nodes, pods, and services, and basic troubleshooting via runbooks.
  • Experience with Ansible or Terraform, participation in continuous integration and continuous delivery such as GitHub Actions, or access and security tooling such as Teleport or Vault.
  • Progress toward relevant Linux, networking, Kubernetes, cloud, or security certifications over time.

About the Company

Nscale

Develops and operates AI data centers

Nscale provides end-to-end AI infrastructure by designing, owning, and operating data centers, GPU compute, networking, storage, and managed AI cloud services for enterprises, governments, and AI developers. It offers a full stack from colocated data centers to high-performance GPUs and networking, plus managed cloud services, so customers can run AI workloads efficiently with sovereign European deployments powered by renewables and US campus expansions. Unlike many peers, Nscale combines complete infrastructure ownership with a focus on sovereign, renewable-powered sites and strong capital partnerships to support large-scale, secure AI workloads. Its goal is to become a leading, scalable AI infrastructure provider that enables governments, enterprises, and AI builders to deploy and run AI applications quickly and securely on sovereign data centers with global capacity.

Company size
1,001-5,000
Company stage
N/A
Total funding
$13.2B
Headquarters
London, United Kingdom
Founded
2024

Recently Posted Jobs

Explore more opportunities at Nscale

See more jobs