Head of AI Infrastructure & HPC

About the role

An ambitious, well‑capitalised technology organisation in Dubai is seeking a Head of AI Infrastructure & HPC to lead the design, delivery and ongoing performance of its advanced GPU‑accelerated compute platforms. This senior position will shape a growing engineering function, own high‑performance GPU environments and ensure large‑scale AI training and inference workloads run efficiently.

Key responsibilities

• Lead design, implementation and expansion of GPU‑accelerated infrastructure for AI and HPC workloads.
• Define road‑maps covering NVIDIA H100, H200, Blackwell, HGX/DGX, high‑speed interconnects, storage and cluster management.
• Oversee deployments of InfiniBand, RoCE, RDMA, NVLink/NVSwitch and distributed networking.
• Establish hardware integration, qualification testing and production‑readiness methodologies.
• Improve efficiency through GPU performance optimisation, capacity planning and reliability engineering.
• Implement cluster scheduling, workload management and container orchestration (Kubernetes, Slurm, Docker).
• Introduce automation frameworks for provisioning, configuration management and operational support.
• Partner with hardware manufacturers, datacentre providers and technology vendors.
• Recruit, develop and manage specialist engineers across GPU systems, Linux, HPC, networking and platform infrastructure.

Required profile

• Extensive experience building, scaling or managing GPU‑intensive infrastructure in AI/HPC environments.
• Proven ability to translate complex infrastructure requirements into reliable, production‑ready systems.
• Demonstrated technical leadership of multidisciplinary engineering teams.

Required skills

• NVIDIA GPU architectures (H100, H200, B200, HGX, DGX).
• High‑bandwidth networking: InfiniBand, Ethernet, RoCE, RDMA.
• GPU communication and performance tools: NCCL, CUDA, NVLink, NVSwitch, DCGM.
• Linux‑based HPC environments and distributed computing.
• Container and scheduling platforms: Kubernetes, Slurm, Docker.
• Server provisioning, hardware qualification and infrastructure automation.
• Scripting/automation: Python, Bash, Go, Ansible.