HPC & AI Cluster Network Architect — Hybrid/Remote

CommonAI CIC; Cambridge; Hybrid; GBP 55,000 - 90,000; Full time

Benefits offered by this job

Collaborative environment

High impact role

Salary & benefits

Professional development

Networking opportunities

Cambridge office

Job summary

CommonAI CIC in Cambridge is seeking an engineer to design and evolve the physical networks connecting clusters in the Scaling Inference Lab (SIL). You will own cabling, switch layouts and fabrics across lab sites in the UK, ensuring robust, scalable interconnects for real workloads.

The role requires deep experience with HPC/AI networks, InfiniBand/RoCE, fibre optics, and clear documentation. You will collaborate with project teams, suppliers and operators to deliver well‑defined designs and

Qualifications

• Experience designing networks for HPC/AI cluster deployments from specification to installation.
• Clear understanding of out-of-band, frontend and accelerator fabric networking.
• Knowledge of InfiniBand/RoCE and similar lossless fabrics.

Responsibilities

• Design the physical network for new clusters: topology, switch placement, cable schedules and BOMs.
• Specify fibre and copper infrastructure in detail and select matching transceivers and media.
• Plan changes to existing clusters to enable re-cabling with minimal disruption.
• Act as technical authority for suppliers: drawings, specifications, quotations, inspections.
• Maintain accurate as-built documentation of racks, cables, ports and links.
• Coordinate with project managers and engineers on bring-up and fault-finding.

Skills

Network design

HPC/AI clustering

Fiber optics

InfiniBand/RoCE

Technical communication

Tools

NetBox

Job description

CommonAI CIC in Cambridge is seeking an engineer to design and evolve the physical networks connecting clusters in the Scaling Inference Lab (SIL). You will own cabling, switch layouts and fabrics across lab sites in the UK, ensuring robust, scalable interconnects for real workloads.

The role requires deep experience with HPC/AI networks, InfiniBand/RoCE, fibre optics, and clear documentation. You will collaborate with project teams, suppliers and operators to deliver well‑defined designs and