AI Infra & Platform Ops Resident: Kubernetes, GPUs, NVAIE

Avensys Consulting; Riyadh; On-site; SAR 350,000 - 520,000; Full time

Job summary

Avensys invites an experienced AI Infrastructure and Platform Operations Resident for Riyadh, KSA. The role focuses on BCM and server lifecycle management, Kubernetes administration, GPU orchestration, and Run:ai resource management to support enterprise AI workloads.

Ideal candidates have 8+ years of hands-on experience with NVAIE, NIM, Triton endpoints, RoCE/RDMA, and storage performance, and can coordinate cross-functional changes while maintaining uptime and incident response.

Qualifications

• 8+ years of experience in IT infrastructure or AI platform operations.
• Experience with Kubernetes and cluster operations.
• Experience with Nvidia AI Enterprise (NVAIE) and NVIDIA NIM.
• Experience with Triton endpoint support and troubleshooting.
• Experience with GPU health, utilization monitoring and AI infrastructure performance.

Responsibilities

• BCM and server lifecycle management.
• Kubernetes administration and cluster operations.
• Run:ai resource pools, quotas, priorities, fair-share, reservations, and queue management.
• GPU health, utilization, capacity, and workload-placement monitoring.
• PowerEdge, PowerScale, AI networking, RoCE/RDMA, and storage performance.
• NVAIE, NIM, Triton endpoint support and troubleshooting.
• Linux, incident management, change control, patching, upgrades, backup, and recovery.
• Maintain runbooks for cluster health, provisioning, workloads, GPU allocation, and recovery.

Skills

NVAIE

NIM

Triton endpoint

Kubernetes

GPU orchestration

PowerEdge

PowerScale/OneFS

AI networking

RoCE/RDMA

storage performance

Run:ai

Job description

Avensys invites an experienced AI Infrastructure and Platform Operations Resident for Riyadh, KSA. The role focuses on BCM and server lifecycle management, Kubernetes administration, GPU orchestration, and Run:ai resource management to support enterprise AI workloads.

Ideal candidates have 8+ years of hands-on experience with NVAIE, NIM, Triton endpoints, RoCE/RDMA, and storage performance, and can coordinate cross-functional changes while maintaining uptime and incident response.