Principal Software Engineer, AI Networking

About the role

NVIDIA is seeking a Principal Software Engineer to lead the transformation of AI networking systems. This role involves managing complex customer engagements, developing product and architecture direction, and influencing NVIDIA's networking technologies.

Responsibilities

• Lead the technical strategy for AI Factory networking deployments at strategic customers, including architecture reviews, risk assessments, and multi-phase execution plans.
• Serve as the principal-level technical authority for embedded networking products like BlueField and ConnectX, covering the surrounding technology ecosystem including DOCA, RDMA, RoCE, and Infiniband.
• Engage deeply with hyperscalers and AI Factory customers through design-in, coding, bring-up, performance tuning, failure analysis, and production hardening.
• Partner with internal engineering, product, and architecture teams to transform customer needs into product features, reference architectures, tooling, and guidelines.
• Drive performance, reliability, and debuggability improvements across customer stacks and translate findings into actionable product, firmware, and software roadmap items.

Requirements

• BS/MS/PhD in Computer Science, Computer Engineering, Electrical Engineering, or equivalent experience.
• 15+ years of relevant industry experience, including technical leadership across complex systems.
• Deep knowledge of networking protocols and distributed systems, with a strong understanding of RoCE/InfiniBand, L1–L4 fundamentals, and performance/latency tradeoffs.
• Proven low-level software expertise with proficiency in C/C++ and comfort debugging across firmware, driver, and user space.
• Demonstrated experience in high-performance networking and system-level debugging, including packet drops, retransmissions, congestion, QoS, ordering, and buffer management.
• Excellent interpersonal skills, with the ability to clearly explain complex topics to engineers, PMs, and customer collaborators, and align cross-organizational teams toward a decision.

Qualifications

• Experience in customer-facing technical leadership at hyperscalers/CSPs/AI factories (or similarly complex production environments).
• Hands-on expertise with DPDK, DOCA, RDMA verbs, NCCL, CUDA-aware networking, congestion control, and performance tuning at scale.
• Experience building internal tools, telemetry, and automation that improve triage speed and operational excellence.
• Demonstrated innovation: patents, publications, hackathons, rapid prototyping, or shipping new architecture/features end-to-end.
• Experience leading multi-team initiatives across geo/time zones, with clear examples of influence without authority as well as eager and proactive in bringing to bear AI-powered tools to accelerate debugging, documentation, and day-to-day engineering efficiency while maintaining strong engineering judgment.

Benefits

We offer competitive salaries and a generous benefits package, including equity and benefits. Applications for this job will be accepted at least until July 10, 2026. This posting is for an existing vacancy. NVIDIA is an equal opportunity employer committed to fostering an inclusive work environment.