[Hiring] Principal Solutions Architect @Nscale

Role Description

We are hiring a Principal Solution Architect to design and deliver end-to-end GPU infrastructure solutions that power some of the world's most demanding AI workloads. This is a highly technical, customer-facing role at the intersection of AI infrastructure, solution architecture, and engineering.

• Translate complex customer requirements into scalable, high-performance infrastructure designs spanning GPU compute, networking, storage, data centre systems, and orchestration platforms.
• Own solution designs end-to-end, from bare-metal GPU fabrics and high-performance networking through to the orchestration layers customers use to run their AI workloads.
• Leverage modern AI tooling and automation to accelerate solution design, architecture validation, and deployment readiness.
• Act as a trusted technical advisor, ensuring our solutions deliver exceptional performance, reliability, and scalability.

Qualifications

• 10+ years of relevant experience in solution architecture, infrastructure engineering, high-performance computing, or large-scale data centre environments.
• Strong experience designing and delivering large-scale GPU infrastructure and AI/HPC solutions.
• Deep technical understanding across the full GPU infrastructure stack, from bare-metal compute and networking through to orchestration and workload management.
• Proven ability to translate complex customer requirements into detailed technical architectures and implementation plans.
• Strong analytical and problem-solving skills, with the ability to troubleshoot complex infrastructure and performance challenges.
• Excellent communication skills, with the ability to explain complex technical concepts to both technical and non-technical audiences.

Requirements

• Strong understanding of GPU architectures, including NVLink/NVSwitch, PCIe, memory bandwidth, and GPUDirect RDMA.
• Deep expertise in high-performance networking, including InfiniBand, RoCE, congestion control, leaf/spine architectures, and rail-optimised fabric design.
• Experience with bare-metal infrastructure provisioning, including PXE/iPXE, firmware management, and BMC technologies.
• Hands-on experience with Kubernetes and Slurm in production environments.
• Understanding of how network topology and GPU interconnect design impact distributed AI workload performance.
• Hands-on experience with NCCL and collective communication benchmarking, including nccl-tests.
• Familiarity with NVIDIA performance validation and diagnostic tooling, including Fabric Manager, DCGM, nvidia-smi, and nvbandwidthtest.
• Experience with InfiniBand and Ethernet fabric diagnostics, including tools such as ibdiagnet.
• Strong scripting and automation skills using Python, Go, or similar languages.
• Experience with Infrastructure-as-Code tools such as Terraform and Ansible.
• Experience with storage technologies supporting AI workloads, including Lustre, GPFS, WEKA, BeeGFS, or large-scale NFS.
• Understanding of distributed training and inference infrastructure requirements.
• Familiarity with AI/ML frameworks such as PyTorch, DeepSpeed, or Megatron-style training stacks.
• Knowledge of checkpointing, data pipeline design, and high-throughput storage architectures.
• Understanding of GPU cluster reliability, burn-in testing, health checks, and observability tooling such as Prometheus, Grafana, and DCGM exporters.

Benefits

• Highly competitive package (base + equity) with reviews every 12 months.
• Opportunity to design and deliver cutting-edge GPU infrastructure powering the next generation of AI.
• Dynamic progression plan tailored to your ambitions.
• Human-First Flexibility: A flexible workplace that trusts Nscalers to deliver, giving you the autonomy to shape your day around life's moments.