GPU AI Infrastructure Architect — HPC Networking

Gateway Search Pte Ltd; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

Gateway Search Pte Ltd is seeking an AI Infrastructure Architect in Singapore to design and deliver GPU infrastructure, including servers, networking, and rack deployment. You will validate hardware, plan connectivity, and support deployment with vendors and data center teams.

The role requires hands-on experience with GPU server architectures and high-speed networking like InfiniBand or RoCE. You will work closely with vendors and engineers to ensure robust, scalable solutions.

Qualifications

• Hands-on experience designing GPU infrastructures and data center deployments.
• Experience selecting GPU servers, CPUs, memory, NICs, and storage for performance and cost.
• Ability to design cluster topology and connectivity plans, including switches and uplinks.

Responsibilities

• Design GPU infrastructure solutions for compute servers, networking, storage, and rack deployment.
• Evaluate GPU servers and components balancing performance, availability, and cost.
• Develop cluster network topology, switch sizing, uplinks, and cabling plans.
• Produce architecture proposals and BOMs with vendors for validation and feasibility.
• Provide guidance during installation, bring-up, and testing; troubleshoot hardware issues.
• Coordinate with data center teams on rack space, power, and cooling requirements.
• Demonstrate strong communication and technical documentation skills.

Skills

GPU infrastructure

Cluster networking

Vendor coordination

Technical documentation

Power/cooling planning

Education

Bachelor's degree in Computer Science or related

Tools

InfiniBand

RoCE

NVLink

NVSwitch

Job description

Gateway Search Pte Ltd is seeking an AI Infrastructure Architect in Singapore to design and deliver GPU infrastructure, including servers, networking, and rack deployment. You will validate hardware, plan connectivity, and support deployment with vendors and data center teams.

The role requires hands-on experience with GPU server architectures and high-speed networking like InfiniBand or RoCE. You will work closely with vendors and engineers to ensure robust, scalable solutions.