Distinguished Engineer - Rack Scale Architecture
Core
Drive end-to-end software architecture for NVIDIA's rack-scale data center systems, including firmware, kernel drivers, OS, networking, and manageability software.
Role type
Distinguished Engineer (System Architecture)
Builds
Rack-scale GPU/HPC systems and associated software stacks for enterprise and cloud providers
Domain
Data Center / High-Performance Computing / Networking
Deliverable
production ML models
Required skills
System architecture and design, SW/HW interface design, Networking protocols (Ethernet, InfiniBand), Accelerator software (GPUs, DPUs, FPGAs), Out-of-band and in-band management, System management protocols (Redfish, IPMI), Security tradeoffs, Strategic planning, Technical mentorship
Preferred skills
Large-scale cloud and cluster deployment, Data center product lifecycle leadership, CXL/UCIE/C2C architectures, Storage and networking technologies
Technologies
NVLink, InfiniBand, Grace CPUs, Redfish, IPMI, CXL, UCIE
Responsibilities
Drive software end-to-end architecture for rack-scale products; Translate forward-looking plans into formal software requirements; Ensure high quality and reliable software; Work directly with major customers to align roadmaps; Work with business partners and vendors to shape products; Develop roadmaps for new technologies and protocols; Mentor architects and engineering teams; Make key technical decisions in ambiguous situations
