NVIDIA
Senior Software Architect - Deep Learning and HPC Communications
$184K–$356K
US, CA, Santa Clara +4 · Senior · Full-time · Posted Oct 13, 2026
- Location
- US, CA, Santa Clara
US, MA, Westford
US, TX, Austin
US, Remote
US, NC, Durham - Experience
- 5+ years of relevant experience.
- Education
- M.S./Ph.D. degree in CS/CE or equivalent experience.
About the job
Designs and evaluates hardware and software communications technologies to improve performance and scalability for deep learning and HPC workloads on next-generation GPU platforms.
What you'll do
- Identify communication bottlenecks and design and implement technologies to accelerate AI and HPC workloads.
- Co-design hardware and software solutions for next-generation platforms with GPU, networking, and software architects.
- Build proofs of concept, run experiments, model performance, and simulate large GPU clusters.
What they ask for
- M.S./Ph.D. in CS/CE or equivalent experience.
- 5+ years of relevant experience.
- Excellent C/C++ programming and debugging skills; experience with parallel programming models and at least one communication runtime.
- Deep understanding of operating systems, computer and system architecture, network architecture, and Linux.
Nice to have
- CUDA programming and NVIDIA GPU experience; knowledge of high-performance networks such as InfiniBand, RoCE, or NVLink.
- Experience with deep learning frameworks such as PyTorch or TensorFlow, deep learning parallelism, or HPC applications.
Pay
184,000 USD - 287,500 USD base salary (Level 4); 224,000 USD - 356,500 USD base salary (Level 5)
Tools
C/C++, MPI, SHMEM, NCCL, NVSHMEM, OpenSHMEM, UCX, UCC
AI focus
Develops communications technology that accelerates AI and deep learning workloads; NVIDIA also states it uses AI tools in recruiting processes.
Model-written summary · the original posting is authoritative