Senior AI Hardware Systems Engineer (GPU/PCIe, Linux)

Nebius Group; United States; On-site; USD 180,000 - 224,000; Full time

Benefits offered by this job

Health insurance

401(k) plan

Parental leave

Remote work reimbursement

Disability & life insurance

Job summary

Nebius Group is seeking a senior systems engineer to diagnose and resolve complex hardware and platform failures in high-performance AI infrastructure. You will work across Linux, GPU servers, PCIe, firmware, power, and cooling to identify true root causes and craft repeatable troubleshooting procedures.

The role requires ownership, curiosity, and the ability to turn ambiguous problems into evidence-based conclusions, enabling reliable AI compute infrastructure for data and model workloads.

Qualifications

• Minimum 5 years of hands-on systems engineering experience across Linux, server hardware, firmware, PCIe, and GPU or high-performance computing platforms.
• Extensive Linux experience for hardware debugging, system-level troubleshooting, and platform investigation.
• Knowledge of the Linux kernel and experience with kernel-level debugging or troubleshooting.
• Strong knowledge of modern server architecture, particularly in high-performance, GPU-based environments.
• Strong knowledge of NVIDIA GPU platforms and diagnostic tooling (nvidia-smi, DCGM, NCCL, NVLink/NVSwitch).
• Deep knowledge of PCIe topology, enumeration, and error handling with a focus on root causes.
• Experience benchmarking systems for performance, stability, power, and thermal behavior.
• Experience with firmware interactions across BIOS/UEFI, BMC, CPLD/FPGA, GPU firmware, and NIC firmware.
• Experience with low-level hardware communication protocols (I2C, SMBus, PMBus).
• Demonstrated ability to troubleshoot complex hardware, software, and networking issues.

Responsibilities

• Form technical hypotheses, design targeted tests, and analyze system logs to isolate failures.
• Diagnose and resolve complex hardware/platform issues across Linux, GPU servers, PCIe, firmware, power and cooling.
• Turn ambiguous problems into evidence-based, repeatable troubleshooting procedures.
• Collaborate across hardware, software, and network teams to determine root causes and remediation steps.

Skills

Linux systems

GPU HPC

Kernel debugging

NVIDIA diagnostics

PCIe debugging

Performance benchmarking

Firmware interfaces

Hardware protocols I2C

Root cause analysis

Python/Go scripting

Tools

nvidia-smi

DCGM

NCCL

NVLink

NVSwitch

Fabric Manager

PCIe tools

BIOS/UEFI tools

Job description

Nebius Group is seeking a senior systems engineer to diagnose and resolve complex hardware and platform failures in high-performance AI infrastructure. You will work across Linux, GPU servers, PCIe, firmware, power, and cooling to identify true root causes and craft repeatable troubleshooting procedures.

The role requires ownership, curiosity, and the ability to turn ambiguous problems into evidence-based conclusions, enabling reliable AI compute infrastructure for data and model workloads.