PrincipalArchitect,AINetworking

An applied research team within NVIDIA’s Networking Systems & Software Architecture group is solving some of AI’s hardest infrastructure problems. The team builds systems-level software that moves data between GPUs, nodes, and storage at the speed modern AI demands—spanning low-level transport optimization, hardware-software co-design, and communication frameworks that plug directly into production AI stacks. The team's charter expands into emerging domains including quantum computing interconnects. This Principal Architect role leads the research agenda and architectural direction for how NVIDIA’s AI systems communicate at scale—across GPUs, DPUs, NICs, and heterogeneous storage. It requires someone who defines project scope from scratch, publishes original work, and translates research breakthroughs into production-grade software that ships industry-wide! **What you will be doing:** * Setting the long-term technical vision for distributed AI communication systems—GPU-to-GPU, GPU-to-storage, and cross-node data movement. * Conducting original research and prototyping next-generation networking solutions over RDMA, NVLink, and GPUDirect. * Driving hardware-software co-optimization with GPU, DPU, NIC, and network switch. Investigating fundamental bottlenecks in communication runtimes for large-scale AI workloads (KV cache transfer, disaggregated prefill/decode, model parallelism). * Integrating networking capabilities into AI serving stacks such as vLLM, SGLang, and TensorRT-LLM. * Publishing findings, representing NVIDIA in industry forums and standards bodies, and mentoring senior engineers across the organization. **What we need to see:** * 15+ years in systems software and/or networking with deep expertise in high-performance networking (InfiniBand, RoCE, RDMA, NVLink), communication libraries (e.g. NIXL, NCCL, UCX, MPI, NVSHMEM), and GPU accelerated systems, with track record of defining and delivering complex, cross-team technical initiatives from research conc