NVIDIA, the pioneer in GPU technology and leader in accelerated computing, is seeking a Distinguished Software Architect to join their GPU Communications Libraries and Networking team. This role represents a unique opportunity to shape the future of deep learning and high-performance computing communications infrastructure.
The position involves working on cutting-edge communication libraries like NCCL, NVSHMEM, and UCX that power applications running on massive GPU clusters. You'll be at the forefront of designing next-generation data center platforms that push the boundaries of what's possible in AI and HPC. The role requires deep expertise in GPU architecture, high-performance networking, and distributed systems.
As a Distinguished Software Architect, you'll be responsible for researching and implementing new communication technologies, co-designing hardware and software solutions with various architectural teams, and ensuring seamless integration across the technology stack. The impact of your work will be direct and significant, as communication performance between GPUs directly affects application performance at scale.
The ideal candidate brings 15+ years of experience, with demonstrated leadership in HPC/DL communications evidenced through patents, publications, and conference presentations. You should be an expert in parallel programming models, GPU architecture, and high-performance networking technologies. Strong programming skills in C/C++ and experience with deep learning frameworks are essential.
This role offers the opportunity to work at the intersection of hardware and software, pushing the boundaries of communication technology in one of the world's most innovative companies. You'll collaborate with leading researchers and engineers while helping to shape the future of AI and HPC infrastructure. The position includes competitive compensation, equity, and the chance to work on technologies that are transforming multiple industries.