Network Engineer, GPUaaS

About the role

Singtel Digital InfraCo's RE:AI division builds Asia's advanced and sustainable AI infrastructure ecosystem, offering GPU-as-a-Service (GPUaaS) data centre networking. As Network Engineer, you will support customers running AI and high-performance computing (HPC) workloads on large-scale distributed GPU clusters, deepening expertise in AI data centre networking and HPC cloud platforms.

Key responsibilities

• Provide operational support for GPU infrastructure services within defined SLAs, including preventive measures to minimise downtime.
• Give technical network support and guidance to users of GPU-accelerated systems.
• Monitor, troubleshoot and improve network performance, introducing automation to detect and resolve issues.
• Manage relationships and expectations with internal and external stakeholders.
• Design, develop and operate high-performance networking solutions for AI and GPUaaS platforms.
• Collaborate on network security initiatives and implement cybersecurity best practices aligned with industry standards and certifications.
• Coordinate with cross-functional teams to deliver network solutions on time.
• Prepare and present technical reports, including network optimisation plans, outage reports and SLA reports.
• Contribute to network benchmarking and stay informed on networking and GPU technology advances.
• Take part in scheduled or on-call support outside normal hours, including nights, weekends and public holidays.

Required profile

• Diploma or degree in Computer Science, Information Technology, Network Engineering or a related field; networking and Linux certifications are advantageous.
• Strong understanding of TCP/IP, VLANs and subnetting.
• Experience with advanced routing protocols and architectures such as VXLAN, EVPN and BGP.
• Hands-on experience with network monitoring, automation and configuration tools.
• English verbal, written and presentation skills; analytical, problem-solving skills and stakeholder management.

Required skills

• Network operating systems and platforms: Cumulus, Arista, Ubuntu, Proxmox, Red Hat.
• Tools: Zabbix, NetBox, Ansible, Python scripting, CI/CD pipelines.
• Desirable: Linux, hypervisors, storage (NFS, object storage), infrastructure-as-code, IaaS/PaaS and NVIDIA GPU platform knowledge.