Senior Manager, Software Development - GPU Accelerated Storage

NVIDIA NVIDIA · Semiconductors · Santa Clara, CA +2

This role is for a Senior Manager, Software Development focused on GPU Accelerated Storage within NVIDIA's data center systems. The primary responsibility is to optimize storage access by enabling direct data paths between GPU memory and storage, bypassing the CPU. The role involves collaborating on reference storage platform designs and leading a global team to drive the storage acceleration roadmap across next-generation systems, frameworks, and applications. It requires deep knowledge of data storage platforms, databases, vector databases, NVMe, RDMA, and system-level architecture, along with significant experience in leading software development teams.

What you'd actually do

  1. Look across the platform, from applications and frameworks to drivers and firmware — to identify software and hardware opportunities that accelerate storage performance. This includes developing new driver features, C++/CUDA libraries, and removing performance & power bottlenecks.
  2. Drive alignment of the storage acceleration roadmap across next-generation systems, frameworks, applications, and the broader storage ecosystem.
  3. Spokesperson for the global team, championing initiatives both internally and externally.
  4. Lead and coordinate the planning, scheduling, and carrying out of team projects and deliverables, ensuring successful completion and accountability for the global team.
  5. Contribute to the planning and execution of NVIDIA’s reference storage platforms.

Skills

Required

  • Deep knowledge of data storage platforms, databases, vector databases.
  • In-depth understanding of NVMe, high performance RDMA network protocols, and related technologies.
  • Deep understanding of system level architecture, such as topologies, interconnects, memory hierarchy, interrupts, and memory-mapped IO.
  • Strong interpersonal, verbal and written communications skills.
  • Successful experience leading team with numerous complex products, competing priorities, and successful delivery on the mission.
  • Bachelor’s or preferably Master’s or Doctoral (Ph.D) degree or equivalent experience in Computer Science, Electrical Engineering, or a related field.
  • 12+ overall years in the industry, including 6+ years growing, mentoring and managing teams with similar responsibilities.

Nice to have

  • Development experience in storage software such as key-value storage, file systems, object storage systems and vector databases.
  • Knowledge of operating system development, including thread and process management, virtual and device memory (e.g., dmabuf), and user-level network and storage I/O.
  • Experience in CUDA programming, exceptional C/C++ programming skills.
  • Previous experience working with system software for accelerators such as GPUs, DPUs, or FPGAs.
  • Internals of frameworks like PyTorch and JAX.

What the JD emphasized

  • 12+ overall years in the industry, including 6+ years growing, mentoring and managing teams with similar responsibilities.