Lead HPC & AI Infrastructure Architect
Core
Lead architecture and technical direction for large-scale CPU and GPU platforms across on-premises, public cloud, and specialized GPU cloud providers for HPC and AI workloads.
Role type
Lead HPC & AI Infrastructure Architect
Builds
Secure, reliable, scalable, and cost-effective platforms for tightly coupled HPC workloads, distributed model training, and inference services.
Domain
High-Performance Computing (HPC) and Artificial Intelligence (AI) Infrastructure
Deliverable
production ML models
Required skills
Large-scale cluster architecture, CPU/GPU topology design, Slurm scheduling, Kubernetes orchestration, RDMA networking, Infrastructure as Code (Terraform/Ansible), Linux troubleshooting, Distributed systems, GPU computing platforms, Observability (Prometheus/Grafana), Security (federated identity, secrets management)
Preferred skills
AMD Instinct GPUs, ROCm, RCCL, Multi-node PyTorch, LLM serving platforms (vLLM/SGLang), Specialized GPU providers (Crusoe/TensorWave), High-performance storage (Lustre/IBM Storage Scale/BeeGFS), Rack power and liquid cooling coordination
Technologies
Slurm, Kubernetes, Terraform, Ansible, Python, Bash, Git, Prometheus, Grafana, ROCm, RCCL, PyTorch, vLLM, SGLang, Lustre, InfiniBand, RoCE
Responsibilities
Lead HPC and AI infrastructure roadmap and reference architectures; Design CPU/NUMA and GPU topology, interconnects, networking, and storage; Architect RDMA-capable fabrics and validate collective communication; Standardize provisioning and fleet upgrades via IaC and CI/CD; Define production acceptance criteria and partner with SRE on fault isolation; Benchmark HPC and distributed AI workloads for performance and cost; Incorporate security controls into platform designs; Lead architecture reviews and mentor engineers.
Seniority
Senior, hands-on IC with leadership responsibilities
