Software Engineering Manager – GPU Communications Libraries
NVIDIA
Opportunity published from Workday · applications stay with the original hiring source.
About the opportunity
- The GPU Communications Libraries and Networking team at NVIDIA delivers communication libraries like NCCL, NVSHMEM, and UCX for Deep Learning and HPC, running on scales up to tens of thousands of GPUs connected via high-speed interconnects and networking.
- Communication performance directly impacts end-to-end application performance, and we are seeking a technical leader to manage our NVSHMEM and UCX libraries.
- This is an opportunity to push the limits of state-of-the-art technology, participate in feature design, and help realize NVIDIA's vision for accelerated computing platforms.
Key responsibilities
- Lead, mentor, and grow the library engineering team, managing project planning, execution, quality, and performance.
- Participate directly in feature design and implementation as part of a technical leadership role.
- Interact with internal and external partners and researchers to understand use cases, requirements, and define product roadmaps.
- Continuously review and identify improvement opportunities in established processes, infrastructure, and practices to ensure efficient and transparent execution.
Qualifications
- 10+ overall years of experience in the software industry with specialization in HPC networking or system software.
- 4+ years of management experience.
- BS, MS, or Ph.D. in CS, CE, EE (related technical field) or equivalent experience.
- Prior systems software, communication runtime, or high-performance networking software development experience with a successful track record of delivering complex software features through the full product life cycle.
- Strong understanding of computer system architecture, operating systems principles, HW-SW interactions, and performance analysis/optimizations.
- Excellent C/C++ programming and debugging skills in Linux.
- Experience balancing multiple projects with competing priorities and communicating effectively across different teams and time zones.
Nice to have
- Experience with parallel programming models (MPI, SHMEM) and communication runtimes (MPI, NCCL, NVSHMEM, OpenSHMEM, UCX, UCC), plus CUDA, OpenMP, OpenACC, or pthreads.
- Background with RDMA, high-performance networking technologies (InfiniBand, RoCE, Ethernet, EFA), network architecture, and network topologies, alongside knowledge of HPC and ML/DL fundamentals.
- Experience with Deep Learning Frameworks such as PyTorch and TensorFlow.
Benefits
- Base salary range of 184,000 USD - 287,500 USD for Level 2, and 224,000 USD - 356,500 USD for Level 3, determined by location, experience, and peer pay.
- Eligibility for equity and comprehensive benefits.
Job details
- Applications accepted at least until October 26, 2026.
- Posting type: Existing vacancy.
About the company
- NVIDIA has been transforming computer graphics, PC gaming, and accelerated computing for more than 25 years through great technology and amazing people.
- Today, NVIDIA taps into the unlimited potential of AI to define the next era of computing, with GPUs acting as the brains of computers, robots, and self-driving cars.
- Recognized as one of the technology world's most desirable employers, NVIDIA offers an innovative environment for creative and autonomous talent.
Application: UpVagas lists this opportunity from the hiring company or its recruitment platform. The application is completed at the original source.
Source transparency. UpVagas organizes information provided by the company or recruitment source. Always review the original listing before submitting personal information.
