Senior Remote GPU Infrastructure Engineer

Nscale; New York (NY); On-site; USD 120,000 - 170,000; Full time

Job summary

Nscale seeks a Senior Infrastructure Support Engineer to own GPU fleet health and high‑performance fabrics. This hands‑on L2/L3 role bridges GPU hardware, Linux, networking, and data centre operations, connecting Support, DC Operations, and Engineering.

You will drive incidents end‑to‑end, perform root‑cause analyses, implement automation, and mentor engineers. Travel to Nscale or customer sites may be required for onsite expertise.

Qualifications

• 6+ years in infrastructure, operations, or support engineering in production environments.
• 2–3+ years hands‑on with GPU, HPC, or large‑scale data centre estates.
• Strong written and spoken communication across customers and stakeholders.
• Deep understanding of GPU drivers, firmware, and runtime stacks for AI clusters.
• Experience with east‑west fabric diagnostics and high‑speed networks.
• Proven ability to author/run changes with risk assessment and backout plans.

Responsibilities

• Join and lead the Support duty rotation as a senior escalation point.
• Diagnose GPU node faults across driver, firmware, and hardware layers.
• Own east‑west fabric health; diagnose and isolate faults in InfiniBand/RoCE fabrics.
• Investigate data-path issues on high‑performance storage platforms.
• Perform structured, hypothesis‑driven investigations and RCA.
• Author and execute changes in live customer environments with proper risk assessment.
• Improve dashboards, alerts, and runbooks; drive automation to reduce toil.
• Record and communicate status with clear customer‑impact notes.
• Design automation scripts and small tools to reduce manual effort.
• Be a primary escalation point; mentor mid‑level engineers and share knowledge.
• Lead with disciplined decision making; travel onsite as needed.

Skills

GPU & HPC expertise

Communication skills

GPU drivers & DCGM

East–west fabric debugging

Linux systems engineering

Networking fundamentals

Observability & incident response

Automation scripting

Change & risk assessment

Mentoring engineers

Leadership under pressure

Travel for onsite work

Tools

nvidia-smi / DCGM tooling

mlxlink / ibdiagnet

Redfish / BMC

MAAS

Pyxis / Enroot

Slurm

Prometheus / Grafana

Terraform / Ansible

Git / GitOps

OpenStack tooling

Job description

Nscale seeks a Senior Infrastructure Support Engineer to own GPU fleet health and high‑performance fabrics. This hands‑on L2/L3 role bridges GPU hardware, Linux, networking, and data centre operations, connecting Support, DC Operations, and Engineering.

You will drive incidents end‑to‑end, perform root‑cause analyses, implement automation, and mentor engineers. Travel to Nscale or customer sites may be required for onsite expertise.