Principal Network Engineer (AI Infrastructure)

Location: UK

Company

Nscale provides GPU cloud infrastructure for AI start-ups and enterprise customers, supporting high-performance AI development and workloads at scale.

What you will do

• Define, design, validate, and evolve InfiniBand, RoCE, and Ethernet fabric architectures across racks, rows, and data centres.
• Set technical direction for BGP, EVPN-VXLAN, LACP, QoS, and high-performance network integration with bare-metal provisioning and cluster management.
• Lead GitOps-based network automation using Python, Ansible, Infrastructure-as-Code, and CI/CD workflows.
• Design perimeter security, firewalls, NAT, VPNs, security policies, and multi-tenant network segmentation.
• Establish SLOs, observability, telemetry, monitoring, alerting, runbooks, and operational standards for network services.
• Lead architecture reviews, complex incidents, technical escalations, cross-functional decisions, and engineer mentoring.

Requirements

• 10+ years of network engineering experience in HPC, AI, hyperscale, or large-scale data centre environments.
• Hands-on RDMA-aware networking experience with InfiniBand and/or RoCE, including OpenSM or NVIDIA UFM.
• Expert knowledge of BGP, EVPN-VXLAN, Clos/spine-leaf architectures, and production platforms such as Cumulus, Nokia, or Arista EOS.
• Strong Python and Ansible automation skills, plus Git workflows and Infrastructure-as-Code or CI/CD tools such as Terraform, GitLab CI, or GitHub Actions.
• Deep experience designing firewall infrastructure with Juniper SRX and/or Palo Alto, as well as telemetry and observability for high-throughput environments.
• Ability to define architecture and technical strategy, lead complex incidents, influence without formal authority, communicate trade-offs, and mentor engineers.

Culture & Benefits

• Collaborative, supportive, and innovation-focused engineering environment.
• Highly competitive base salary and equity package with reviews every 12 months.
• Flexible workplace with autonomy over daily scheduling.
• Progression plan aligned with individual ambitions and opportunities to influence a global AI platform.
• Inclusive environment supporting applications from underrepresented groups and workplace accommodations.