AI Infrastructure Hardware Support Engineer-Onsite

Role Purpose:

Provide hands-on hardware troubleshooting and support for NVIDIA-based AI/HPC infrastructure, including A100, H100, B200, B300, GB200 and GB300 platforms.

Key Responsibilities:

• Troubleshoot GPU servers, AI clusters, GPUs, CPUs, memory, system boards, NVLink/NVSwitch, NICs, DPUs, storage, power and cooling.
• Analyse BMC logs, NVIDIA Xid/ECC errors and hardware telemetry using NVIDIA SMI, DCGM, IPMI and OEM tools.
• Troubleshoot InfiniBand, Ethernet, RoCE, cables and transceivers.
• Perform hardware health checks, component replacement, testing and post-repair validation.
• Resolve BIOS, firmware, BMC, driver and hardware compatibility issues.
• Coordinate hardware replacements and escalations with OEMs/NVIDIA.
• Maintain incident records, troubleshooting guides and RCA reports.
• Support shift-based/on-call requirements when needed.

Required Skills:

• 3+ years of Data Centre, Server or HPC hardware support experience.
• Strong enterprise server and component-level troubleshooting skills.
• Knowledge of NVIDIA GPUs and multi-GPU systems.
• Working knowledge of Linux and server-management tools.
• Understanding of networking, storage, rack power and cooling.
• GPU/AI infrastructure experience is highly preferred.
• Strong troubleshooting, fault isolation and incident-management skills.