Wilson Hiring! Full Time AI Infrastructure Solution Architect, GPU in - Ricebowl

Our client, a fast-growing AI infrastructure and high-performance computing company building and operating large-scale GPU compute environments to power next-generation AI workload, is looking for an experienced AI Infrastructure Architect to take ownership of the physical infrastructure behind large-scale AI and GPU compute environments. This is anewly created and hands-on architecture role focused on GPU server hardware, high-speed network fabrics and physical-layer networking—from architecture and hardware selection through rack-level implementation.

Responsibilities:

• You will own the end-to-end hardware architecture for GPU clusters and AI data centers, covering compute servers, GPU/heterogeneous accelerators, storage, networking, racks and supporting infrastructure.
• A major part of the role will involve designing high-speed GPU fabrics using InfiniBand and/or RoCE, including NVLink/NVSwitch architectures, Fat-Tree/Spine-Leaf topologies, switch and port planning, NICs, optical transceivers/modules, DAC/AOC and fiber/cabling strategy.
• You will lead hardware selection and compatibility validation, develop detailed BOMs, and provide technical oversight during rack-and-stack, structured cabling, power-on, hardware acceptance and physical-layer troubleshooting.
• You will also help shape future architecture around next-generation GPU/NPU platforms, high-density infrastructure and liquid cooling.

Requirements:

• Ideal candidates must have proven experience in HPC, GPU infrastructure, AI data centers or large-scale compute environments.
• You should bring deep knowledge of server architecture—including CPU/GPU, memory, PCIe, HGX-class platforms, NVLink and NVSwitch—combined with hands-on expertise in InfiniBand/RoCE, RDMA, data-center switches, optical transceivers, DAC/AOC, fiber and physical network design.
• Experience independently architecting and delivering GPU clusters, including hardware selection, capacity planning, BOM development, rack layouts, power and cooling considerations, is important.
• Candidates who have designed or deployed environments ranging from hundreds to thousands of GPUs will be particularly relevant.

If you’ve built the infrastructure behind large-scale GPU and AI compute and are ready to take ownership of what powers the next generation of AI, we want to hear from you.

Interested? Here’s what to do next:

Please apply to the role with a recent copy of your CV. We will be in touch once we review your profile for suitability.

___

Singapore Employment Agency Personnel No.: R1109150

Singapore Employment Agency License No: 16S8069

Privacy Statement: By submitting an application and sharing your personal information with us, you consent to such information being collected, held, used, and disclosed in accordance with our Privacy Policy (*************-policy)