Network Engineer (GPU/HPC)

About the role

The Infrastructure team manages the Group's Internet infrastructure and business applications, covering data centers, connectivity, cloud, networking, systems, storage and security. Network Engineers deploy the network infrastructure supporting the high bandwidth and low latency needs of the products and services, keeping the data center network efficient, robust, scalable and secure.

Key responsibilities

• Lead end-to-end network architecture design for GPU, HPC and storage clusters, ensuring scalability, resiliency and performance
• Develop High-Level Design (HLD) documentation, including architecture diagrams, design specifications and Bills of Materials (BOM)
• Deploy, configure and support the large-scale production network infrastructure
• Work with TPM and Cloud Platform Engineering teams to accelerate GPU cluster bring-up and ensure timely readiness
• Maintain network documentation, policies, procedures and service records
• Analyse network traffic and capacity requirements, plan future expansion and optimise routing for low latency

Required profile

• Bachelor's degree or higher in Computer Networking, Telecommunications or a related field
• At least 5 years of hands-on experience with Arista, Cisco or Nvidia products
• Able to manage multiple tasks and projects with minimal supervision and to work independently in a fast-paced environment
• Good communication and coordination skills with internal and external stakeholders
• Occasional on-call support required; candidates can only apply once per group

Required skills

• InfiniBand (IB) and RoCE network architecture design
• Protocols such as BGP, OSPF, MPLS, EVPN, Segment Routing, VxLAN, ECMP, ECN, PFC, QoS, IPSec, NAT and IPv6
• Scripting or automation for data center network deployment
• Performance validation, interoperability and troubleshooting, acceptance testing
• Familiarity with compute and storage networking is a plus; CCNP/NCP-IB or equivalent is a plus