Member of Technical Staff, Data Infrastructure
Core
Build a unified, high-performance data layer spanning parallel file systems, object storage, distributed caches, and key-value stores across multiple clouds to ensure GPUs never wait on data for training and inference.
Role type
Senior IC data infrastructure engineer (storage/caching systems)
Builds
Unified data layer for GPU workloads (datasets, checkpoints, model weights, KV cache)
Domain
Cloud infrastructure + AI/ML data pipelines
Deliverable
infrastructure
Required skills
Distributed storage systems design, Caching architecture, Linux kernel internals, High-performance networking (RDMA/RoCE), Kubernetes CSI, Infrastructure as code, Performance profiling
Preferred skills
Core contributor to open-source storage systems, High-performance computing clusters, Petabyte-scale data movement, GPUDirect Storage
Technologies
C/C++, Rust, Go, Python, Kubernetes, Terraform, Pulumi, fio, iperf3, NCCL, eBPF, WEKA, VAST, Lustre, Ceph, MinIO, Slurm, AWS, GCP
Responsibilities
Design and ship distributed storage systems, build distributed/tiered caches, tune OS-level I/O paths, configure high-performance networking, deploy storage clusters across clouds, profile and resolve performance bottlenecks
Seniority
Senior, hands-on IC