GPU HPC Architect

Radiant; Greater London; On-site; GBP 110,000 - 180,000; Full time

Benefits offered by this job

25 days annual leave

Health & Wellbeing: private medical

Cycle to Work Scheme

Gympass subscription

Participation in company shares

Enhanced parental pay & leave

Job summary

Radiant is building purpose-built AI infrastructure with GPU-native platforms and sovereignty at scale. The Cluster Architect will lead low-level designs for high-density AI supercomputers, translating high-level requirements into production-grade deployments with a focus on performance, reliability and cost-efficiency.

You will work with datacenter engineering, software teams, and vendors to deliver scalable HPC/AI deployments, documenting architecture, BOMs, and runbooks for handover to

Qualifications

• BSc in Computer Science, Electrical/Computer Engineering, Physics, Mathematics, or related field.
• 5+ years of solution architecture experience driving design and delivery.
• Deep expertise in datacenter engineering, GPU systems, InfiniBand networking, storage.
• Proficiency with OS-level aspects, Linux kernels, GPUs, NICs, and infra software.
• Experience producing detailed infrastructure designs, BOMs, power/cooling planning.

Responsibilities

• Translate inputs from stakeholders into Radiant’s reference architecture and low-level designs.
• Lead low-level design of high-density AI supercomputer systems including compute, interconnect, storage.
• Participate in Architecture Review Board to ensure collaboration and alignment.
• Evaluate tradeoffs between performance, power, cooling, serviceability, and cost.
• Drive detailed low-level designs with vendors and draft rack layouts, floor plans, BoMs.
• Oversee RFx responses, ensuring designs meet vendor capabilities and constraints.
• Hands-on role in deployment, cabling, benchmarking, and production-grade implementation.
• Establish acceptance criteria for quality, reliability, safety, and performance.
• Collaborate with Software/Solutions teams to support AI workload requirements.
• Define milestones, risks, and deliverables with project management.

Skills

Solution architecture

Datacenter engineering

GPU systems

InfiniBand networking

Linux / kernel drivers

BoM creation

Power & cooling planning

Vendor engagement

Stakeholder communication

NVIDIA CUDA / GPU development

Education

BSc in Computer Science or related engineering

Tools

Kubernetes

NVIDIA CUDA / SDKs

Job description

Radiant is building purpose-built AI infrastructure with GPU-native platforms and sovereignty at scale. The Cluster Architect will lead low-level designs for high-density AI supercomputers, translating high-level requirements into production-grade deployments with a focus on performance, reliability and cost-efficiency.

You will work with datacenter engineering, software teams, and vendors to deliver scalable HPC/AI deployments, documenting architecture, BOMs, and runbooks for handover to