Principal Architect, AI Networking

Details

NVIDIA has been a pioneer in computer graphics, PC gaming, and accelerated computing for over 25 years. The company is now focusing on leveraging AI to usher in the next era of computing, where GPUs power intelligent machines such as robots and self-driving cars.

Responsibilities

• Accelerate NVIDIA Dynamo - KV cache management and large-scale inference.
• Research and develop innovative networking technologies to scale AI networks.
• Co-design software and hardware networking solutions across networking domains, from network transports to AI frameworks.
• Collaborate with NVIDIA's hardware and software architecture and research teams to create cutting-edge networking solutions.
• Lead prototype development to optimize AI training and inference infrastructure.

Requirements

• Master's or Ph.D. in Computer Science, Electrical or Computer Engineering, or equivalent experience.
• 12+ years of relevant academic or industry experience.
• Deep understanding of AI workloads (primarily inference and training) and their network infrastructure impact.
• Strong expertise in Machine Learning/Deep Learning fundamentals, inference runtimes, and frameworks.
• Skilled in C or C++ for systems software development; familiarity with Rust is a plus.
• Curiosity and passion for building leading-edge technology.
• Effective communication and collaboration skills across diverse teams and time zones.

Ways to stand out

• Proven research record.
• Experience with large language model (LLM) inference and AI network/storage requirements.
• Background in storage optimization including file systems, object stores, caches, and coherency.
• Excellent communication abilities.

Benefits

NVIDIA is renowned as a top technology employer offering highly competitive salaries and comprehensive benefits. Additional equity and benefits packages are available.

This position is designated hybrid work (#LI-Hybrid).

Base salary range: 272,000 USD - 425,500 USD (location and experience dependent).