Technical Lead - GPU Infrastructure

About the role

This hands‑on technical leadership position is responsible for architecting and delivering a large‑scale GPU infrastructure platform in Saudi Arabia. You will guide the transition from managed Kubernetes workloads to a bare‑metal GPU environment that supports research, model training, and managed inference.

Key responsibilities

• Own end‑to‑end platform architecture, including high‑level and low‑level designs, technical reviews, and documentation.
• Lead and line‑manage a distributed engineering team covering backend, frontend, DevOps, QA, and documentation.
• Design, build, and operate a managed Slurm service for research and model‑training workloads.
• Oversee GPU infrastructure operations on bare‑metal, including NVIDIA drivers, CUDA, Fabric Manager, NVSwitch, DCGM, and MIG.
• Bootstrap and maintain Kubernetes clusters on bare metal using NVIDIA GPU Operator and Network Operator.
• Define and implement managed inference architecture with multi‑GPU, multi‑node parallelism, autoscaling, and confidential‑compute capabilities.
• Establish observability across control plane, GPU fleet, and applications through metrics, logging, alerting, and SLOs.
• Lead incident response and post‑incident reviews.

Required profile

• Proven experience in technical leadership and team management.
• Strong background in large‑scale GPU compute platforms.
• Ability to make deep technical decisions while overseeing delivery.
• Excellent communication with infrastructure partners, hardware providers, and internal consumers.

Required skills

• GPU infrastructure and bare‑metal provisioning.
• Slurm workload manager.
• Kubernetes, NVIDIA GPU Operator, Network Operator.
• NVIDIA drivers, CUDA, Fabric Manager, NVSwitch, DCGM, MIG.
• KubeVirt, VFIO for GPU isolation.
• Observability tools: metrics, logging, alerting, SLO definition.
• Managed inference design, multi‑GPU and multi‑node scaling, autoscaling, request routing, confidential‑compute.