Sr. GPU Cloud Storage Solutions Expert (SRE SME)
Core
Own the IO layer for AI training/inference, deploying high-performance storage systems and feeding AIOps signals to predict storage faults before they stall training runs.
Role type
Senior Site Reliability Engineer (Storage SRE) for AI GPU Cloud
Builds
High-performance storage layer (WEKA, VAST Data, Ceph, DDN/Lustre) for AI training and inference across US data centers
Domain
AI/ML Infrastructure, High-Performance Computing (HPC), Cloud Storage
Deliverable
production ML models
Required skills
Enterprise/HPC storage operations, AI/ML workload support, WEKA/VAST Data/Ceph/DDN/Lustre, AI training I/O patterns, high-performance storage networking, GPU Direct Storage, storage performance benchmarking, multi-tenant storage isolation, Linux systems internals, runbook-as-code
Preferred skills
AIOps substrate instrumentation, anomaly detection for storage telemetry, automation of SOPs
Technologies
WEKA, VAST Data, Ceph, DDN/Lustre, NFS over RDMA, NVMe-oF, Nvidia CMX, fio, IOR, mdtest, GB200 clusters
Responsibilities
Deploy and operate parallel/distributed storage systems optimized for AI workloads; Implement multi-tenant storage isolation with QoS and access controls; Diagnose and tune storage performance using benchmarking tools; Instrument storage telemetry into the metrics/logs/traces store; Partner with platform team to define storage-fault predictors; Convert incidents into automation (runbook-as-code)
Seniority
Senior, hands-on IC