نوع العقددوام كامل
طبيعة الوظيفةبالموقع
الموقعالرياض

وصف الوظيفة

About the Role

Acuative Middle East is seeking an NVIDIA AI Infrastructure & High Performance Networking Resident Engineer to support a strategic customer data center in Riyadh, Saudi Arabia. This full-time role focuses on NVIDIA GPU-based AI and High-Performance Computing (HPC) infrastructure, with a primary emphasis on high-performance networking technologies. The successful candidate will serve as the primary on-site technical resource, ensuring the availability, reliability, and optimal performance of the infrastructure.

Key Responsibilities

  • Deploy, configure, monitor, and troubleshoot NVIDIA InfiniBand and high-speed Ethernet/RoCE networks.
  • Provide support for NVIDIA ConnectX NICs, Quantum InfiniBand switches, and Spectrum Ethernet switches.
  • Troubleshoot issues related to RDMA, GPUDirect, network connectivity, latency, packet loss, and overall performance.
  • Support 200G, 400G, and 800G networking components, including fiber, optics, DAC, and AOC cabling.
  • Troubleshoot NVIDIA GPU servers, HGX platforms, NVLink, NVSwitch, and cluster connectivity.
  • Monitor and support Direct Liquid Cooling (DLC) systems, including CDUs, pumps, temperature, pressure, and flow.
  • Perform infrastructure health checks, performance testing, preventive maintenance, and incident management.
  • Conduct root-cause analysis and coordinate technical escalations with Supermicro and NVIDIA engineering teams.
  • Maintain technical documentation, network topology, and operational reports.
  • Collaborate directly with customer engineering and data center operations teams.

Required Experience

Candidates should possess 5 to 10 years of experience in supporting NVIDIA GPU-based AI and HPC infrastructure, with a strong background in high-performance networking. This includes practical experience with NVIDIA InfiniBand and Ethernet/RoCE technologies.

Technical Skills

  • Proficiency in deploying and managing NVIDIA InfiniBand and high-speed Ethernet/RoCE networks.
  • Experience with NVIDIA ConnectX NICs, Quantum InfiniBand, and Spectrum Ethernet switches.
  • Troubleshooting expertise in RDMA, GPUDirect, network connectivity, latency, and packet loss.
  • Familiarity with 200G, 400G, and 800G networking components and cabling.
  • Knowledge of NVIDIA GPU servers, HGX platforms, NVLink, and NVSwitch.
  • Understanding of Direct Liquid Cooling systems operations and maintenance.

Work Location

This full-time position is based in Riyadh, Saudi Arabia, requiring on-site presence at a strategic customer data center.

Application Process

Salary information for this role will be disclosed during the interview process.


متطلبات الوظيفة

  • تتطلب ٥-١٠ سنوات خبرة

وظائف مشابهة