GPU Server & Systems Operations Engineer

Nexo Global Inc

Irvine, CA, United States

8 days ago

Role details

Contract type

Permanent contract

Employment type

Full-time (> 32 hours)

Experience level

Experienced

Experience required

4 years minimum

Compensation

$65,000.0 - $90,000.0

Working hours

Regular working hours

Job source

www.indeed.com

Tech stack

Intelligent Platform Management Interface BIOS Computer Clusters Nvidia CUDA Linux RAID Hard Disk Drives Firmware Monitoring of Systems Networking Basics Prometheus Server Administration

+8 more

Shell Script Zabbix Graphics Processing Unit (GPU) Grafana Containerization Kubernetes Slurm Hardware Infrastructure

Job description

Conduct acceptance, rack installation, cabling, initialization and asset registration upon arrival of GPU servers.
Troubleshoot hardware faults involving GPUs, memory, hard drives, power supplies, network adapters, optical modules and other components.
Administer BMC, IPMI, BIOS, RAID, firmware and fundamental Linux operating systems.
Install and maintain foundational environments including NVIDIA drivers, CUDA and DCGM.
Collaborate with the Shenzhen team to resolve failures related to Slurm, Kubernetes, storage and GPU clusters.
Manage server monitoring alerts, routine inspections, spare parts inventory and vendor RMA processes.
Assist the datacenter with power supply, temperature and liquid cooling alerts; compile operation & maintenance SOPs and conduct fault RCA.
Participate in emergency on-call rotation and on-site incident resolution.

Requirements

Minimum 4 years of working experience in server, IDC or Linux systems operations.
Familiar with GPU server hardware, BMC/IPMI, BIOS, RAID and firmware upgrades.
Proficient in Linux, Shell scripting and basic networking knowledge.
Working knowledge of NVIDIA GPU, CUDA, DCGM, NVLink/NVSwitch.
Familiar with monitoring tools such as Prometheus, Grafana and Zabbix.
Experience with Slurm, Kubernetes or container platforms is preferred.
Hands-on experience with B300/GB200/H100/H200 or liquid-cooled servers is preferred.
English communication skills for interactions with datacenter staff and vendors; must possess valid legal work authorization in the United States.