Senior HPC Support Engineer: Linux, Kubernetes, On-Call

Lambda; United States; On-site; USD 122,000 - 162,000; Full time

Benefits offered by this job

Health coverage

Dental

Vision

Wellness stipend

Commuter stipend

401k matched

Flexible PTO

Job summary

Lambda, The Superintelligence Cloud, is a leader in AI cloud infrastructure providing services to researchers, enterprises, and hyperscalers. This role is based in the United States and participates in an on-call rotation to handle the most challenging infrastructure and platform issues.

You will act as a senior escalation point, differentiate among hardware, driver, and kernel problems, and drive permanent fixes by collaborating with engineering and documenting solutions.

Qualifications

• 3+ years in HPC administration, support, or engineering.
• Strong Linux system administration experience.
• Experience with Kubernetes and/or Slurm for cluster orchestration.

Responsibilities

• Serve as senior escalation point for infra issues down to hardware/kernel level.
• Differentiate hardware, driver, kernel, and workload misconfig to resolve issues quickly.
• Identify gaps in processes, tooling, and docs and fix them.
• Build scripts or small tools using AI-assisted techniques to close gaps.
• Perform root-cause analysis across distributed systems and GPU clusters.
• Document solutions and contribute to support procedures.

Skills

HPC administration

Linux admin

Kubernetes

Slurm

CI/CD

AI tooling

Monitoring tools

Kernel debugging

CUDA/NVLink

Networking IB/RoCE

Distributed AI/ML

Cloud networking

Tools

Docker

Terraform

Ansible

Prometheus

Grafana

Datadog

NVIDIA GPU tools

Job description

Lambda, The Superintelligence Cloud, is a leader in AI cloud infrastructure providing services to researchers, enterprises, and hyperscalers. This role is based in the United States and participates in an on-call rotation to handle the most challenging infrastructure and platform issues.

You will act as a senior escalation point, differentiate among hardware, driver, and kernel problems, and drive permanent fixes by collaborating with engineering and documenting solutions.