AI Infrastructure Engineer: HPC GPU Clusters

Veeda AI; Seattle (WA); On-site; USD 180,000 - 240,000; Full time

Job summary

Veeda AI is building the next generation of multimodal foundation world models for Physical AI. The Member of Technical Staff - AI Infrastructure will design, deploy, and operate high-performance GPU clusters running Slurm and Kubernetes.

You will own interconnect fabric, storage pipelines, and observability, ensuring reliability for large-scale distributed DL workloads and real-time data streaming.

Qualifications

• Bachelor's degree in CS/CE or equivalent HPC/infrastructure experience.
• Hands-on Linux HPC cluster administration (Slurm or Kubernetes).
• Strong low-level Linux networking, kernel/PCIe/NUMA tuning, hardware diagnostics.
• Proficiency with automation and IaC tools; code-driven cluster configuration.
• Experience with distributed DL workloads (PyTorch/NCCL, Ray, DeepSpeed).

Responsibilities

• Design, deploy, and operate bare-metal GPU clusters with Slurm and Kubernetes.
• Tune Slurm topology, fair-share, QoS, and preemption for long runs.
• Own the interconnect fabric (InfiniBand, RoCEv2) and validate with nccl-tests.
• Run high-throughput storage (Lustre, WEKA, Ceph) with NVMe tiers.
• Build telemetry pipelines (Prometheus, Grafana, DCGM) to monitor hardware health.

Skills

HPC cluster administration

Slurm / Kubernetes

Networking / kernel tuning

Automation / IaC

Distributed DL workloads

Education

Bachelor's degree in CS/CE or equivalent HPC experience

Tools

Ansible

Terraform

Helm

Python

NFS/Lustre/Ceph

Job description

Veeda AI is building the next generation of multimodal foundation world models for Physical AI. The Member of Technical Staff - AI Infrastructure will design, deploy, and operate high-performance GPU clusters running Slurm and Kubernetes.

You will own interconnect fabric, storage pipelines, and observability, ensuring reliability for large-scale distributed DL workloads and real-time data streaming.