Principal GPU Infra Architect for AI workloads

Nscale; Dublin; On-site; EUR 150,000 - 190,000; Full time

Job summary

Nscale is hiring a Principal Solution Architect to design end-to-end GPU infrastructure for demanding AI workloads. You will architect clusters, storage, and networking, and lead technical engagements with customers to deliver scalable, reliable solutions.

Role requires deep GPU and HPC expertise, plus experience with Kubernetes, Slurm, Terraform, and Ansible. You will collaborate with engineering teams to ensure feasibility and performance, while integrating automation and AI-assisted design

Qualifications

• 10+ years of relevant experience in solution architecture, infrastructure engineering, HPC, or large-scale data centres.
• Deep understanding of GPU infrastructure stacks from bare-metal to orchestration.
• Proven ability to translate complex customer requirements into architectures and plans.
• Strong analytical and problem-solving skills for troubleshooting complex infra.

Responsibilities

• Design end-to-end GPU infrastructure solutions tailored to customer requirements and AI workloads.
• Architect GPU cluster configurations, including rack layouts, power and cooling, and NVLink/NVSwitch domains.
• Design high-performance network architectures (InfiniBand, RoCE, Ethernet) for scalable AI workloads.
• Define storage architectures and data pipelines for large-scale AI training and inference.
• Design orchestration layers (bare-metal provisioning, Kubernetes, Slurm) and multi-tenant scheduling.
• Develop monitoring, alerts, and lifecycle management for GPU clusters.

Skills

Solution architecture

Infrastructure engineering

HPC

GPU infrastructure

Networking

Kubernetes

Slurm

Terraform

Ansible

Python

Tools

Kubernetes

Slurm

Terraform

Ansible

Python

Go

Job description

Nscale is hiring a Principal Solution Architect to design end-to-end GPU infrastructure for demanding AI workloads. You will architect clusters, storage, and networking, and lead technical engagements with customers to deliver scalable, reliable solutions.

Role requires deep GPU and HPC expertise, plus experience with Kubernetes, Slurm, Terraform, and Ansible. You will collaborate with engineering teams to ensure feasibility and performance, while integrating automation and AI-assisted design