Senior Network Architect for AI-Scale Cloud

Nscale; United Kingdom; On-site; GBP 110,000 - 160,000; Full time

Job summary

Nscale, a GPU cloud engineered for AI, is hiring a Staff Network Engineer to lead design, validation and operation of high-performance networking for our AI-focused cloud. You will own InfiniBand/Ethernet fabrics, drive automation via Python/Ansible, implement secure perimeter infrastructure, develop observability, and mentor engineers across sites while partnering with operations and vendors.

This role requires 10+ years of experience in HPC/data center networks, strong leadership, and a

Qualifications

• 10+ years of network engineering experience in HPC/AI/large-scale data centers.
• Hands-on RDMA networking for AI/HPC; InfiniBand/RoCE; OpenSM/UFM.
• Expert data-center routing: BGP, EVPN-VXLAN; Clos/spine-leaf; platforms: Cumulus, Nokia, Arista EOS.
• Strong network automation: Python, Ansible; Git workflows; Terraform; CI/CD.
• Firewall platforms: Juniper SRX and Palo Alto; security policy, HA, multi-tenant segmentation.
• Network telemetry and observability for high-throughput environments.
• Lead complex decisions across networking, systems, storage, HPC/AI; balance performance and reliability.
• Architecture standards and technical strategy beyond a single project; influence engineers.
• Strong communication and mentoring; explain trade-offs to leaders and ops.
• Hands-on, ownership-driven in fast-paced ML-scale infra.

Responsibilities

• Define, design, validate, and evolve large-scale InfiniBand/RoCE and Ethernet fabric architectures.
• Own technical direction for high-performance Ethernet fabrics; BGP, EVPN-VXLAN; reference architectures across sites.
• Design perimeter and security infrastructure — firewalls, NAT, VPN, and security policy architecture.
• Lead network automation strategy in a GitOps model; Python/Ansible tooling for provisioning and validation; CI/CD.

Skills

Network engineering

RDMA networking

InfiniBand

RoCE

OpenSM/UFM

Fabric orchestration

Data center routing

BGP

EVPN-VXLAN

Clos/spine-leaf

Cumulus

Nokia

Arista EOS

Python

Ansible

Terraform

GitLab CI

GitHub Actions

Juniper SRX

Palo Alto

Security policy

HA design

Multi-tenant segmentation

Telemetry

Observability

Technical leadership

Architecture standards

Influence

Mentoring

Hands-on

Tools

Cumulus

Nokia

Arista EOS

Job description

Nscale, a GPU cloud engineered for AI, is hiring a Staff Network Engineer to lead design, validation and operation of high-performance networking for our AI-focused cloud. You will own InfiniBand/Ethernet fabrics, drive automation via Python/Ansible, implement secure perimeter infrastructure, develop observability, and mentor engineers across sites while partnering with operations and vendors.

This role requires 10+ years of experience in HPC/data center networks, strong leadership, and a