Senior Software Engineer, Fabric Networking - GPU
Core
Build and sustain system software enabling efficient communication between GPUs for next-generation AI, deep learning, and high-performance computing platforms.
Role type
Senior IC software engineer (GPU fabric networking)
Builds
System software for GPU-to-GPU communication, next-gen communication hardware/software, and rack-scale NVLink-based systems
Domain
AI infrastructure, High-Performance Computing (HPC), GPU networking
Deliverable
production ML models | product features | infrastructure
Required skills
C/C++ programming, systems programming, Linux kernel internals, multi-core/multi-threaded application development, networking fundamentals (TCP/IP, Ethernet, InfiniBand, RDMA/RoCE), device driver collaboration, shell scripting
Preferred skills
Python, Perl, NVIDIA GPU systems (NVLink, NVSwitch, CUDA), large-scale AI system architecture, server management technologies, software quality assurance techniques, AI-assisted development tools
Technologies
Linux, Windows, KVM, QEMU, Hyper-V, NVLink, NVSwitch, CUDA, InfiniBand, RDMA/RoCE
Responsibilities
Build and roll out system software for GPU communication; Participate in architecture of next-gen communication hardware/software; Drive platform bring-up, feature enablement, and end-to-end validation; Triage and resolve complex software/firmware/networking issues; Contribute to software quality, performance optimization, and scalability; Collaborate with cross-functional teams to deliver software solutions
Seniority
Senior, hands-on IC
