Senior Software Advanced Developer
Core
Develop and prototype advancements in distributed training and inference using NVIDIA's Spectrum-X AI fabric to improve AI app-networking connection and large-scale AI system efficiency.
Role type
Senior IC software engineer (AI networking & distributed systems)
Builds
Production AI networking features, communication libraries, drivers, and firmware integrations for NVIDIA's AI infrastructure.
Domain
AI infrastructure, high-performance networking, distributed computing
Deliverable
production ML models
Required skills
Networking internals (NCCL, RDMA/RoCE, congestion control), Low-level programming (C/C++, kernel, drivers), HW/SW/FW integration, Distributed training systems (PyTorch DDP, Megatron-LM, DeepSpeed)
Preferred skills
Programmable data planes (P4, eBPF, DOCA SDK), NIC firmware scheduling, In-network compute, Open-source contributions, AI factory architectures
Technologies
Spectrum-X, NCCL, RDMA, RoCE, C, C++, PyTorch, Megatron-LM, DeepSpeed, P4, eBPF, DOCA SDK
Responsibilities
Prototype end-to-end solutions for distributed training and disaggregated inference, Analyze and optimize communication flows across application, transport, and network layers, Develop system software spanning communication libraries, drivers, and firmware integrations, Collaborate with hardware, firmware, and SDK teams to co-design network features, Validate and integrate prototypes into NVIDIA's AI infrastructure and products
Seniority
Senior, hands-on IC
