Senior Storage Software Engineer - DGX Cloud

Details

NVIDIA's DGX Cloud Storage team builds storage platforms that support tens of thousands of accelerators, maintain exabytes of data, and power large-scale AI workloads across cloud, neocloud, and on-premises environments. The role involves foundational storage engineering for AI infrastructure as a hands-on individual contributor and technical lead.

Responsibilities

• Contribute code, fixes, and features to open-source parallel and distributed file systems and distributed object storage, while engaging with upstream communities and maintainers.
• Write and review production code and examine kernel, NFS, NVMe-oF, or SPDK source code to diagnose bugs.
• Make technical decisions for storage deliveries against measurable performance and durability targets.
• Triage, troubleshoot, and determine the root cause of storage issues across GPU clusters with tens of thousands of GPUs, including I/O and metadata performance problems, data corruption, and recovery incidents.
• Validate storage architecture, capabilities, performance, and durability through scale tests, benchmarks, recovery drills, and qualification of new builds.
• Define configuration, tuning, and operational best practices for high-performance file systems on GPU infrastructure.
• Help operators and internal customers apply storage configuration and tuning standards.
• Collaborate with training, inference, accelerated-computing, site-reliability, operations, networking, and security teams, as well as cloud providers, neocloud operators, and storage vendors.
• Use modern AI coding and agentic tools for development, debugging, validation, and operations.

Requirements

• BS, MS, or PhD in Computer Science, Electrical Engineering, or a related field, or equivalent experience.
• More than 12 years of direct storage software engineering experience, including extensive experience with a high-performance parallel or distributed file system operating at multi-petabyte scale.
• Contributions to open-source projects involving distributed or parallel file systems.
• Hands-on experience writing and reviewing production code, examining file system, kernel, NVMe-oF, or SPDK source code, and conducting scale tests or recovery drills.
• Experience diagnosing and resolving storage issues in large GPU or HPC clusters, including I/O and metadata performance analysis.
• Strong proficiency in at least one systems language: C, C++, Rust, or Go.
• Proficiency in Python.
• Familiarity with Linux kernel storage and networking stacks, including the block layer, RDMA/RoCE/InfiniBand, NVMe, page cache, VFS, and multipath.
• Understanding of object storage such as S3 or Swift-class systems and block storage such as NVMe-oF and iSCSI.
• Strong written and verbal communication skills, with the ability to explain complex technical trade-offs to engineers, SREs, vendors, and internal customers.
• Comfort working in a 24/7 production environment where storage incidents directly affect GPU availability, with a security-first approach.
• 100% hands-on engineering, including personally writing and reviewing production code, investigating source code, and running scale tests or recovery drills.

Preferred Qualifications

• Maintainer status or sustained contributions to widely used public projects.
• Experience designing or operating storage for AI training or inference at very large GPU scale, with measurable improvements in GPU utilization or reductions in I/O bottlenecks.
• Kernel and file system development experience, including metadata scalability, data placement, failure recovery, or HSM or equivalent technologies.
• Kubernetes and CSI driver development for storage.
• Hands-on experience with SPDK, libfabric, or FUSE performance optimization.

Benefits

• Equity and benefits are provided.
• NVIDIA is an equal opportunity employer committed to an inclusive work environment.