Software Architect, AI Networking

NVIDIA seeks a highly motivated Software Architect to join our Networking Software Architecture team in Israel. In this role, you will work on innovative technologies shaping the future of AI cloud systems and RISC-V networking accelerators. We develop RDMA transport protocols and low-level software within the Networking team. Efficient and fast communication between GPUs directly impacts end-to-end AI application performance an impact that continues to grow with the increasing scale of next-generation systems. In this role you will help design infrastructure, build proofs-of-concept, and evaluate protocol performance to deliver platforms the world has never seen before.

Design & Software Development : Help build and maintain C/C++ infrastructure, embedded software components, and runtime environments supporting programmable RDMA protocols for the East/West (E/W) AI networking stack. PoC & Quantitative Modelling: Build proofs-of-concept, conduct hands-on experiments, and perform quantitative modelling to evaluate performance and drive new networking innovations. Cross-Functional Collaboration: Partner with hardware, firmware, architecture, compiler, and SDK teams to lead cross-stack innovations and integrate scalable networking solutions. Documentation & Technical Clarity: Craft, refine, and maintain technical design documents, test results, and specs to effectively engage with a globally distributed engineering team. What We Need to See Education : Bachelor’s or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent experience. Experience : 0-5 years of experience in low-level software development, networking, or systems programming. Core Skills: Strong proficiency in C and/or C++ and solid system-level debugging skills. Networking & System Foundations: Background or coursework in computer architecture, operating systems, and networking protocols (e.g., TCP/IP, RDMA, or InfiniBand fundamentals). Communication : Strong articulation skills for crafting clear technical documentation and collaborating across global engineering teams. Mindset : Curiosity, strong problem-solving abilities, and an eagerness to constantly learn and improve your technical expertise. Ways to Stand Out from the Crowd Hands-on experience or academic projects focusing on RDMA, InfiniBand, SmartNICs, or packet processing. Familiarity with RISC-V execution environments, RTOS, board bring-up, or driver development. Practical experience with Linux kernel development, network simulation tools, or performance profiling.