Member of Technical Staff - Storage Infrastructure

Role overview Build and operate storage systems that provide dependable, high-throughput access to datasets, checkpoints, and model artifacts for demanding AI workloads. The role balances performance, durability, availability, and cost as GPU clusters and shared research workflows grow.

Responsibilities - Design storage architectures for training data, checkpointing, inference artifacts, and shared workflows. - Deploy and tune parallel file systems, object storage, and local NVMe caching. - Benchmark throughput, latency, metadata performance, and concurrent access using representative workloads. - Improve reliable data access for large-scale compute workloads. - Balance storage performance and availability with durability and operating cost.

Requirements - Experience designing or operating storage infrastructure for demanding workloads. - Familiarity with parallel file systems, object storage, or local caching systems. - Ability to benchmark storage behavior across throughput, latency, metadata, and concurrency. - Strong operational judgment around reliability, durability, and performance.

Benefits and work setup - Full-time role based in San Francisco or remote. - Cash compensation of $150,000–$300,000, plus equity incentives.