Staff Network Architect AI HPC Fabric

Nscale; Dublin; On-site; EUR 120,000 - 160,000; Full time

Job summary

Nscale is hiring a Staff Network Engineer to design and operate the AI-optimized network fabrics that power our GPU cloud. You will own critical network domains end-to-end, drive architecture across low-latency, high-bandwidth networks, and mentor a growing team of engineers across multiple sites.

You will combine hands-on engineering with architectural leadership, defining reference architectures, enforcing standards, and partnering with deployment, data center operations, platform engineering,

Qualifications

• Over 10 years of network engineering experience in HPC/AI or large-scale data centers.
• Hands-on RDMA networking for AI/HPC workloads (InfiniBand/RoCE), subnet managers (OpenSM/UFM).
• Deep knowledge of data center routing and control planes (BGP, EVPN-VXLAN, Clos).
• Strong automation experience using Python and Ansible; IaC with Terraform; Git-based workflows.
• Experience with firewall platforms (Juniper/Nokia/Palo Alto) and security architecture.
• Experience designing telemetry and observability for high-throughput environments.
• Proven ability to lead decisions across multiple teams with influence rather than authority.
• Strong communication and mentoring skills for cross-functional partners.
• Hands-on, ownership-driven in fast-paced infra for ML scale-out.

Responsibilities

• Define and evolve InfiniBand/RoCE and Ethernet fabrics at data-center scale.
• Drive high-performance fabric direction and standardization across sites.
• Design perimeter security architecture, firewalls, NAT, VPN, and policies.
• Lead network automation strategy using GitOps, Python/Ansible tooling, and CI/CD pipelines.
• Guide complex escalations and root-cause analysis to reduce toil and improve SLOs.
• Establish network observability, telemetry, monitoring, and alerting across the fabric.
• Ensure accurate network inventory data and controlled change management.
• Collaborate with deployment, data center operations, and vendors on site delivery.
• Mentor engineers through architecture reviews and incident leadership.
• Identify architectural gaps and implement durable, scalable solutions.

Skills

HPC networking

RoCE/InfiniBand

BGP EVPN-VXLAN

Python Ansible

Terraform GitOps

Observability telemetry

Firewall design

Mentoring

Multi-site leadership

Tools

Cumulus

Nokia

Arista EOS

OpenSM/UFM

Job description

Nscale is hiring a Staff Network Engineer to design and operate the AI-optimized network fabrics that power our GPU cloud. You will own critical network domains end-to-end, drive architecture across low-latency, high-bandwidth networks, and mentor a growing team of engineers across multiple sites.

You will combine hands-on engineering with architectural leadership, defining reference architectures, enforcing standards, and partnering with deployment, data center operations, platform engineering,