Senior Network Engineer

Trulyyy; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

Trulyyy is seeking an experienced AI Cloud Network Operations Engineer in Singapore to operate and optimize high-performance AI/HPC network infrastructure across data centers. The role emphasizes incident handling, performance troubleshooting, and scalable network configuration for large GPU clusters.

Responsibilities include AI network operations, fabric optimization with InfiniBand/RoCEv2, and automation using Python/Go; familiarity with Prometheus/Grafana/Zabbix is expected.

Qualifications

• 5+ years of Network Operations / Network Engineering experience in large-scale Cloud, Internet, Data Center, Carrier or HPC environments.
• Hands-on knowledge of BGP, OSPF, VXLAN, EVPN and ECMP with independent troubleshooting on enterprise networking equipment.
• Exposure to InfiniBand and/or RoCEv2 with understanding of PFC and ECN for high-performance workloads.
• Experience with monitoring tools such as Prometheus, Grafana, Zabbix and related platforms.

Responsibilities

• AI Network Operations — Monitor and operate large-scale data center and AI/HPC networks across switches, routers, links, bandwidth and health.
• Incident & Performance Troubleshooting — Own network incidents and troubleshoot latency, jitter, packet loss and NCCL throughput issues.
• Network Configuration & Changes — Execute and troubleshoot BGP, OSPF, VXLAN, EVPN and ECMP environments, VLANs and bandwidth changes.
• AI/HPC Fabric Operations — Support GPU cluster networking using InfiniBand and/or RoCEv2 with lossless Ethernet tech like PFC and ECN.
• Monitoring & Automation — Improve visibility and efficiency using monitoring platforms and Python/Go-based automation.

Skills

Network Operations

BGP

OSPF

VXLAN

EVPN

ECMP

InfiniBand

RoCEv2

NVIDIA Spectrum/Quantum

Prometheus

Grafana

Zabbix

Python

Go

Tools

Prometheus

Grafana

Zabbix

NVIDIA Spectrum/Quantum

Arista

Cisco

Job description

Our client is a global technology company operating large-scale AI/HPC and data center infrastructure across multiple international markets. As its AI cloud capabilities continue to scale, the company is looking for an experienced AI Cloud Network Operations Engineer to operate and optimize high-performance network infrastructure supporting large-scale GPU computing environments.

Job Responsibilities

• AI Network Operations — Monitor and operate large-scale data center and AI/HPC networks across switches, routers, optical links, bandwidth utilization and network health.
• Incident & Performance Troubleshooting — Own network incidents and troubleshoot complex performance issues including latency/jitter, packet loss, GPU-to-GPU communication and NCCL throughput degradation.
• Network Configuration & Changes — Execute and troubleshoot BGP, OSPF, VXLAN, EVPN and ECMP environments, including VLAN, routing, traffic isolation and bandwidth changes.
• AI/HPC Fabric Operations — Support high-performance GPU cluster networking using InfiniBand and/or RoCEv2, including lossless Ethernet technologies such as PFC and ECN.
• Monitoring & Automation — Improve network visibility, operational efficiency and reliability using monitoring platforms and Python/Go-based network automation.

Job Requirements

• 5+ years of Network Operations / Network Engineering experience within large-scale Cloud, Internet, Data Center, Carrier or HPC environments.
• Strong hands-on knowledge of BGP, OSPF, VXLAN, EVPN and ECMP, with independent troubleshooting capability on enterprise networking equipment.
• Practical exposure to InfiniBand and/or RoCEv2, with understanding of PFC, ECN and lossless networking for high-performance workloads.
• Experience operating large-scale infrastructure using vendors such as NVIDIA Spectrum/Quantum, Arista or Cisco, together with monitoring tools such as Prometheus, Grafana, Zabbix or equivalent.
• Exposure to large-scale GPU clusters / AI infrastructure is highly advantageous, particularly NVIDIA H100/GB200 environments, NCCL/MPI, NVIDIA UFM, network automation or optical/DCI networks.