Director, Rack-Scale Software Architecture

Details

NVIDIA has been transforming computer graphics, PC gaming, and accelerated computing for more than 25 years. Today, NVIDIA is using AI to define the next era of computing, with GPUs powering computers, robots, and self-driving cars.

NVIDIA has a rapidly expanding ecosystem of data center platform and node designs, ranging from single-node HGX/DGX systems to large, multi-node NVLink domain rack architectures. These designs support NVIDIA's enterprise and cloud provider businesses by combining NVIDIA GPUs, NVLink, InfiniBand networking, Grace CPUs, and an optimized AI and HPC software stack.

The team is seeking a highly technical and motivated manager to lead the team responsible for rack-scale system software architecture, spanning firmware, kernel drivers, operating systems, networking, fabrics, user-mode drivers, and manageability software. The role involves working with internal component leads and industry-leading hyperscalers and cloud service providers to bring these products to market.

Responsibilities

• Drive the end-to-end software architecture for NVIDIA's rack-scale products.
• Maintain a deep understanding of the product portfolio and roadmap, translating forward-looking plans into clear, formal software requirements that anchor execution across the organization.
• Ensure high-quality, reliable software and serve as a trusted architectural partner to teams requiring guidance or oversight.
• Work directly with major customers to understand their requirements and align their roadmaps with NVIDIA's roadmap.
• Present the team vision to senior NVIDIA leaders and external leaders using strong communication skills.
• Provide technical leadership and career mentorship to the team.
• Make key technical decisions in ambiguous situations.

Requirements

• Bachelor's or master's degree in Computer Engineering, Computer Science, or a related discipline, or equivalent experience.
• At least 15 years of overall experience in system architecture and design, including at least 8 years of proven management experience.
• Deep experience designing architectures for scalable and performant server systems, particularly at the software and hardware interface.
• Proven leadership skills and strong ownership of projects involving large-scale, sophisticated codebases.
• Previous experience working with complex system software for accelerators such as GPUs, DPUs, or FPGAs.
• Strong managerial, problem-solving, and critical-thinking skills.
• Ability to operate comfortably in highly matrixed organizations while holding a leadership position.
• Strong interpersonal, verbal, and written communication skills.

Preferred Qualifications

• Knowledge of large-scale cloud and cluster-level deployment and management systems.
• Experience designing robust, resilient, and performant scale-up fabrics.
• A demonstrated track record of leading data center products across their entire lifecycle, including inception, pre-silicon development, post-silicon bring-up, manufacturing, and deployment.
• Strong understanding of networking technologies and protocols, such as Ethernet and InfiniBand.
• Familiarity with CXL, UCIe, and other chip-to-chip technology architectures.
• Knowledge of storage and networking technologies.

Benefits

The role includes eligibility for equity and benefits.