Principal Solutions Architect

You will design end-to-end GPU infrastructure for AI workloads, translate customer requirements into technical architectures, and advise on trade-offs. You will validate GPU cluster performance, troubleshoot infrastructure issues, automate design and deployment workflows, and define storage, monitoring, and operational-readiness requirements.

Responsibilities

• Design end-to-end GPU infrastructure solutions
• Architect GPU clusters, network fabrics, storage, and orchestration layers
• Translate customer requirements into technical architectures and reference designs
• Advise customers on infrastructure design decisions and trade-offs
• Plan deployments and validate performance, reliability, and operational requirements
• Execute GPU-cluster acceptance and validation plans
• Benchmark GPU performance and troubleshoot multi-node cluster issues
• Automate architecture, bills of materials, network topologies, and deployment artefacts
• Use infrastructure-as-code and scripting to streamline deployment
• Define monitoring, observability, health-check, and operational-readiness requirements
• Research emerging GPU, networking, storage, and AI infrastructure technologies

Requirements

• 10+ years of relevant solution architecture, infrastructure engineering, HPC, or large-scale data centre experience
• Experience designing large-scale GPU infrastructure and AI or HPC solutions
• Knowledge of GPU architecture, networking, bare-metal provisioning, Kubernetes, and Slurm
• Experience with NCCL and GPU performance validation tools
• Experience troubleshooting multi-node GPU clusters
• Python or Go scripting and automation skills
• Experience with Terraform and Ansible
• Knowledge of AI workload storage technologies
• Knowledge of PyTorch, DeepSpeed, or Megatron-style training stacks

Benefits

• Equity package
• Compensation reviews every 12 months
• Flexible work arrangements