Senior System Software Architect, AI and GPU Networking
Core
Design and lead architectural and development efforts for AI networking, GPU networking, and AI data center technologies to optimize AI workloads and systems communication.
Role type
Senior System Software Architect (AI/GPU Networking)
Builds
GPU networking offerings (NVIDIA Dynamo, NIXL, UCX), runtime systems, communication libraries, and AI-specific technologies.
Domain
AI Systems, High Performance Computing (HPC), Software Defined Networking (SDN), Accelerated Computing
Deliverable
production ML models
Required skills
System architecture, algorithm design, system programming, computer architecture, operating systems, virtualization, networking, storage, performance profiling, optimization, hardware feature definition, software development
Preferred skills
Research track record, Deep Learning frameworks (NCCL, UCX, MPI), Inference and Training workload optimization (Prefill/Decode, data parallelism, Tensor parallelism, FSDP), CPU/GPU/memory/storage/networking expertise
Technologies
NVIDIA Dynamo, NVIDIA NIXL, NVIDIA UCX, NCCL, UCX, MPI, FSDP
Responsibilities
Enhance GPU Networking offerings for AI workloads; Co-design hardware features for data movement; Identify and evaluate new technologies and partner relationships; Lead architecture and design of new technologies; Lead proof-of-concept development
Seniority
Senior, hands-on IC