Senior Cloud Network Operations Engineer

Straitdeer Pte. Ltd.; Cyberjaya; On-site; MYR 180,000 - 280,000; Full time

Job summary

Straitdeer Pte. Ltd. in Malaysia is seeking a Senior Cloud Network Operations Engineer to monitor a full network estate and drive incident response for AI-oriented environments.

You will work with architecture and delivery teams to resolve microsecond-level jitter, optimize NCCL throughput, and implement change requests in real time while maintaining strict SLAs.

Qualifications

• 5+ years of network operations experience in large-scale environments.
• Degree in CS, Telecoms, Electronics or related field.
• Strong hands-on knowledge of BGP, OSPF, VXLAN, EVPN, ECMP.
• Comfortable with CLI on enterprise-grade switches/routers from NVIDIA Spectrum, Arista or Cisco.
• Knowledge of InfiniBand or RoCEv2 and lossless congestion control (PFC, ECN).
• Familiar with monitoring tools such as Zabbix, Prometheus, Grafana, Cacti, MTR.
• Fluent in Chinese and English.

Responsibilities

• Monitor the full network estate 24x7 and track health, bandwidth, and load.
• Own incident response for outages and failures, meeting SLAs.
• Handle real-time network change requests (traffic isolation, VLANs, routing).
• Collaborate with architecture and delivery teams on high-performance AI networks.
• Manage ticket lifecycle from creation to resolution.
• Produce regular network health reports and post-incident reviews.

Skills

BGP/OSPF

VXLAN/EVPN/ECMP

Troubleshooting CLI

AI network signals

Monitoring tooling

Education

CS/Telecoms/Electronics degree

Tools

NVIDIA Spectrum

Arista

Cisco

InfiniBand

RoCEv2

Zabbix

Prometheus

Grafana

Cacti

MTR

Job description

Senior Cloud Network Operations Engineer

Monitor the full network estate 24x7 — switches, routers, optical transport — tracking device health, link bandwidth, and traffic load in real time. Work alongside architecture and delivery teams to hunt down the hard stuff: microsecond-level jitter affecting GPU-to-GPU communication, NCCL throughput degradation, and other performance issues unique to high-performance AI networks. Act as the technical interface for internal R&D teams and key customers, handling network change requests in real time: traffic isolation, VLAN adjustments, routing policy changes, temporary bandwidth upgrades, and more.

Key responsibilities

Monitor the full network estate 24x7 — switches, routers, optical transport — tracking device health, link bandwidth, and traffic load in real time

Keep a close eye on AI-specific network signals: the indicators that hint at a problem before it becomes one

Own incident response for network outages, link failures, and device-down events — moving fast, closing the loop, and hitting SLA targets every time

Work alongside architecture and delivery teams to hunt down the hard stuff: microsecond-level jitter affecting GPU-to-GPU communication, NCCL throughput degradation, and other performance issues unique to high-performance AI networks

Manage the full ticket lifecycle for routine network requests — fault reports, access requests, resource allocation — from creation through escalation to resolution

Produce regular network health reports and post-incident reviews; build out the team's knowledge base so every hard-won lesson is captured

Act as the technical interface for internal R&D teams and key customers, handling network change requests in real time: traffic isolation, VLAN adjustments, routing policy changes, temporary bandwidth upgrades, and more

Execute changes under a rigorous change management process — writing the change plan, running the implementation, and keeping the blast radius on AI training jobs as close to zero as possible

About you

5+ years of network operations experience in large-scale internet, cloud, carrier NOC, or datacenter environments

Degree in CS, Telecoms, Electronics, or similar

Strong hands-on knowledge of BGP, OSPF, VXLAN, EVPN, and ECMP — both configuration and the underlying forwarding mechanics

Comfortable at the CLI of enterprise-grade switches and routers from vendors like NVIDIA Spectrum/Quantum, Arista, or Cisco; able to troubleshoot independently

Working knowledge of InfiniBand or RoCEv2, and a solid grasp of lossless network congestion control — PFC, ECN, and why they matter for AI traffic

Familiar with network monitoring tools: Zabbix, Prometheus, Grafana, Cacti, MTR, or equivalent

Fluent in Chinese and English