Senior Infrastructure Support Engineer — Remote-First + Equity

Nscale; United States; Remote; USD 120,000 - 170,000; Full time

Benefits offered by this job

Highly competitive package with base +

Equity and bonus programs

Remote-first team

Flexible workplace

Job summary

Nscale is building an AI cloud with hands-on responsibility for GPU fleets, Linux operations, and high‑performance networks. You’ll own escalations, drive incidents, and improve reliability across data-center estates, collaborating with Engineering, DC Ops, and Support.

The role demands 6+ years in infra, 2–3+ years with GPU/HPC, and a strong focus on automation, incident response, and customer impact. Remote-first team with potential onsite work as needed.

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.
• Able to explain complex technical detail clearly, in tickets, incident updates, and to stakeholders at all levels.
• Practical, current experience with GPU drivers, firmware, and runtime stacks on AI training and inference clusters.
• Hands-on experience with RDMA fabrics (InfiniBand/RoCE); diagnose faults and perform fault isolation.
• Slurm operations for large multi-GPU jobs; containers/MPI; diagnose queues and job failures.
• Strong Linux systems engineering; troubleshoot across compute, storage, and network.
• Experience with BMC/Redfish, firmware management, bare-metal provisioning (MAAS).
• Solid networking fundamentals: L2/L3, routing, VLANs, VXLAN.
• Observability and incident response with dashboards (Prometheus/Grafana).
• SRE-style operations; write/runbooks; automate to reduce toil.
• Automation and Git; Bash, Python; Ansible/Terraform.
• Data centre fundamentals; leadership and mentorship.
• Adaptability; travel for onsite work.

Responsibilities

• Join Support duty rotation as senior escalation point across incidents and changes.
• Diagnose and remediate GPU node faults across driver, firmware, and hardware layers.
• Own east–west fabric health; diagnose link faults and validate topology across InfiniBand/RoCE.
• Investigate data-path issues on storage platforms and storage–network interactions.
• Lead structured investigations; drive root-cause analysis and long-term fixes.
• Author and execute changes in live customer environments with risk assessments and backouts.
• Improve dashboards, alerts, and runbooks to prevent repeats; document patterns.
• Record, update, and resolve tickets with clear customer-impact notes.
• Design automation scripts and small tools to reduce toil.
• Mentor mid-level engineers; support knowledge sharing and training.
• Travel to Nscale or customer sites for onsite expertise as needed.

Skills

GPU

HPC

Linux

Data centers

Automation

SRE

Incident response

Leadership

Tools

nvidia-smi

DCGM

Redfish

MAAS

Job description

Nscale is building an AI cloud with hands-on responsibility for GPU fleets, Linux operations, and high‑performance networks. You’ll own escalations, drive incidents, and improve reliability across data-center estates, collaborating with Engineering, DC Ops, and Support.

The role demands 6+ years in infra, 2–3+ years with GPU/HPC, and a strong focus on automation, incident response, and customer impact. Remote-first team with potential onsite work as needed.