Customer Engineer, High Performance Computing, Google Cloud
Core
Pre-sales solutions architect for Google Cloud TPU/HPC infrastructure, designing architectures and proving concepts for large-scale AI/ML workloads.
Role type
Senior IC customer engineer (HPC/TPU)
Builds
Pre-sales technical strategies, proofs of concept, and deployment architectures for Google Cloud TPU solutions.
Domain
Cloud computing, High Performance Computing (HPC), AI/ML infrastructure
Deliverable
production ML models
Required skills
HPC infrastructure architecture, distributed AI/ML workload orchestration, Kubernetes/Slurm, Linux host software, cluster control planes, data center infrastructure integration, performance profiling, kernel optimization
Preferred skills
Large-scale accelerator cluster integration, fleet management APIs, IaaS operations tools, hardware management (BMC/Redfish), Prometheus telemetry, liquid cooling/OCS/ToR networking, hardware RMA workflows
Technologies
PyTorch, JAX, vLLM, SGLang, Kubernetes, Slurm, XProf, PyTorch Profiler, Pallas, MaxText, TPU, Optical Circuit Switches
Responsibilities
Drive pre-sales technical strategy and build go-to-market solutions for large-scale TPU agreements; Serve as trusted technical advisor on end-to-end architecture and integration strategies; Guide customers through technical qualification, pod layout design, system turn-up, and operational readiness; Conduct workload performance profiling, migration, and optimization for TPU hardware topologies.
