Cluster Architect
Radiant · Remote · London ·
COMPENSATION
€•• – ••k
Listed on the employer's job board (Ashby)
- Senior Cluster Architect
- GPU & HPC Focus
- Fully Remote
- AI Infrastructure
- Vendor Management
As a Cluster Architect at Radiant, you will drive the technical strategy and reference architecture for GPU compute platforms. You translate high-level customer requirements into operational supercomputers, leading low-level design for HPC and AI infrastructure. Working with Datacenter Engineering and vendors, you ensure scalable, secure, and efficient solutions for demanding environments.
What you'll do
- Translate stakeholder inputs into reference architecture and low-level design blueprints
- Lead low-level design of high-density AI supercomputer systems including compute and storage
- Evaluate tradeoffs between performance, power, cooling, serviceability, and cost
- Drive creation of detailed designs such as rack layouts, floor plans, and Bill of Materials
- Manage technical evaluations and vendor responses during RFx processes
- Oversee manufacturing, cabling, configuration, and benchmarking of new deployments
- Produce architecture diagrams, runbooks, and documentation for operations handover
- Collaborate with software teams to ensure infrastructure supports AI workload requirements
What you'll need
- BSc in Computer Science, Electrical/Computer Engineering, Physics, or related field
- 5+ years of solution architecture experience driving technical design and delivery
- Deep expertise in GPU systems, high performance networking, and storage architecture
- Proficiency in Linux kernel drivers, GPUs, NICs, and infrastructure software solutions
- Experience producing detailed infrastructure designs including network topologies and BoMs
- Understanding of datacenter fit-out, mechanical/electrical constraints, and failure modes
- Excellent communication skills for interfacing with engineering teams and vendors
- Comfortable working in fast-moving, high-complexity technical environments
Nice to have
- Experience in HPC, Telco, or hyperscale supercomputing environments
- Knowledge of networking fundamentals, TCP/IP stack, and data center designs
- Familiarity with monitoring, telemetry, and observability for infrastructure health
- Expertise in cloud orchestration like Kubernetes, Docker Swarm, or Slurm
- Hands-on experience with NVIDIA systems, CUDA, RoCE, InfiniBand, or ARM CPUs
- Proficiency in C/C++ programming, parallel programming, and GPU development
Stack
- GPU Systems
- InfiniBand
- Linux
- Kubernetes
- Docker
- Slurm
- CUDA
- C/C++
What you get
- 25 days of annual leave
- Private medical insurance via Bupa
- Gympass subscription for gyms and wellbeing apps
- Company shares program participation
- Enhanced parental pay and leave
- Dedicated learning time for skill development
- Cycle to Work Scheme support