Principal Architect, AI Networking

Job Description

Assume the pivotal role of Principal Architect, driving the future of distributed AI communication systems. Engage in innovative research, prototyping groundbreaking networking solutions to propel NVIDIA's AI capabilities. Lead hardware-software co-optimization, tackle fundamental bottlenecks in AI workloads, and integrate advanced networking into AI serving stacks. Share your insights through publications and industry forums. Mentor senior engineers, shaping the organization's technical expertise in AI networking.

Qualifications

1. Possess 15+ years of experience in systems software and/or networking. 2. Demonstrate deep expertise in high-performance networking (InfiniBand, RoCE, RDMA, NVLink). 3. Have strong knowledge of communication libraries (e.g. NIXL, NCCL, UCX, MPI, NVSHMEM). 4. Showcase experience with GPU-accelerated systems. 5. Have a proven track record of leading complex technical initiatives. 6. Hold an MS, PhD, or equivalent in Computer Science, or a related field. 7. Show deep understanding of computer architecture and OS-level networking. 8. Understand ML systems concepts, including transformer architectures and distributed training. 9. Be proficient in programming languages such as C, C++, Rust, and Python.

Benefits

- Competitive salaries - Comprehensive benefits package - Eligibility for equity