Principal Solutions Architect

nscaleoperationsukltd; United States; Hybrid; USD 180,000 - 260,000; Full time

Benefits offered by this job

Equity

Base pay + equity

Annual compensation reviews

Progression plan

Job summary

nscaleoperationsukltd is seeking an experienced Infrastructure Architect to design and deliver end-to-end GPU infrastructure for demanding AI workloads. You will translate customer requirements into scalable architectures spanning compute, networking, storage, and orchestration, collaborating with customers and engineering teams to ensure deployment readiness.

You will plan high-performance GPU clusters, configure interconnects, and validate performance through multi-node benchmarking, applying

Qualifications

• Experience designing GPU/HPC infrastructure for AI workloads.
• Ability to translate customer needs into scalable designs and architectures.
• Familiarity with high-speed networking and GPU interconnects (InfiniBand, RoCE, Ethernet).
• Experience with orchestration and provisioning tools such as Kubernetes or Slurm.
• Ability to validate multi-node performance and troubleshoot issues.

Responsibilities

• Design GPU clusters, including rack layouts, power and cooling, and GPU interconnect configurations.
• Architect high-performance network fabrics, storage, data pipelines, and orchestration for AI training and inference.
• Turn customer requirements into detailed architectures, specifications, and reference designs; advise on technical trade-offs.
• Coordinate with engineering and deployment teams to confirm feasibility and meet performance, reliability, and operational needs.
• Plan and lead cluster acceptance tests and performance benchmarking; investigate issues across multi-node systems.
• Use automation and AI tools to accelerate architecture development, validation, bills of materials, and deployment preparation.

Skills

GPU cluster design

HPC infrastructure

Kubernetes

Slurm

NCCL benchmarking

Networking interconnects

Tools

Prometheus

Grafana

DCGM

Job description

Role overview

Design and deliver end-to-end GPU infrastructure for demanding AI workloads, translating customer needs into scalable technical solutions. The role spans compute, networking, storage, data centre design, orchestration, performance validation, and deployment readiness, with close collaboration across customers and engineering teams.

Responsibilities

• Design GPU clusters, including rack layouts, power and cooling, and GPU interconnect configurations.
• Architect high-performance network fabrics, storage, data pipelines, and orchestration for AI training and inference.
• Turn customer requirements into detailed architectures, specifications, and reference designs; advise on technical trade-offs.
• Coordinate with engineering and deployment teams to confirm feasibility and meet performance, reliability, and operational needs.
• Plan and lead cluster acceptance tests and performance benchmarking; investigate issues across multi-node systems.
• Use automation and AI tools to accelerate architecture development, validation, bills of materials, and deployment preparation.

Requirements

• Strong experience designing GPU, HPC, hyperscale cloud, or AI infrastructure solutions.
• Knowledge of GPU cluster design, high-speed interconnects, and networking such as InfiniBand, Ethernet, or RoCE.
• Experience with infrastructure orchestration and provisioning, such as Kubernetes, Slurm, or bare-metal systems.
• Ability to validate GPU cluster performance, including benchmarking with NCCL or comparable tools, and troubleshoot multi-node issues.
• Strong customer-facing communication and the ability to turn workload needs into practical technical designs.
• Collaborative, accountable approach to complex work in a fast-moving environment.

Nice to have

• Experience in solution architecture, presales engineering, or technical advisory work; familiarity with observability and GPU health tooling such as Prometheus, Grafana, or DCGM; and experience applying AI-assisted engineering or infrastructure automation.

Benefits and work setup

The listing describes a flexible workplace, a compensation package combining base pay and equity, annual compensation reviews, and a progression plan.