Senior Storage Software Engineer - DGX Cloud
Core
Build and maintain high-performance parallel and distributed file systems and object storage to support tens of thousands of GPUs for AI training and inference workloads.
Role type
Senior Storage Software Engineer (Technical Lead)
Builds
Open-source parallel/distributed file systems, distributed object storage, and GPU cluster storage infrastructure.
Domain
AI/ML Infrastructure, High-Performance Computing (HPC), Distributed Systems
Deliverable
production ML models
Required skills
C, C++, Rust, Go, Python, Linux kernel storage stacks, NVMe-oF, RDMA/RoCE/InfiniBand, S3/Swift, iSCSI, SPDK, libfabric, FUSE
Preferred skills
Kubernetes CSI driver development, HSM, metadata scalability, failure recovery, AI training/inference storage optimization
Technologies
SPDK, libfabric, FUSE, NVMe, RDMA, RoCE, InfiniBand, S3, Swift, Kubernetes, CSI
Responsibilities
Contribute code to open-source parallel and distributed file systems; Triage and root-cause large-scale storage issues across GPU clusters; Validate storage architecture and run scale tests/benchmarks; Define configuration and tuning guidelines for high-performance file systems; Partner with SRE, networking, and cloud teams on common architecture.
Seniority
Senior, hands-on IC with technical lead responsibilities
