Senior Switch Software Researcher

NVIDIA NVIDIA · Semiconductors · Raanana, Israel +2

Research role focused on Ethernet Networks and Switch ASICs for high-performance ML/AI computing, involving research into switch-level capabilities, hardware/software tradeoffs, ML methods for network understanding and control, and building PoCs. The role also involves evaluating emerging protocols and collaborating with architects and advanced development teams.

What you'd actually do

  1. Research switch-level capabilities to cluster-scale outcomes, including interactions with orchestration, job scheduling, collective communication patterns, and observability.
  2. Explore hardware/software tradeoffs for high-radix Ethernet switches aimed at next-generation AI fabrics.
  3. Investigate ML methods for traffic understanding, control-plane assistance, configuration optimization, and other high-leverage areas.
  4. Build and implement Proof-of-Concepts (PoCs) using network/system simulations, performance models, and data-driven analyses to compare design alternatives to influence architecture decisions.
  5. Continuously evaluate emerging protocols, software stacks, open-source initiatives, partner ecosystems, and academic work for alignment with the Ethernet switch roadmap.

Skills

Required

  • M.Sc. or Ph.D. in Computer Science, Computer Engineering, or Electrical Engineering
  • 12+ years of experience in system architecture, networking technologies, and embedded software, including 3+ years functioning as a Researcher in that fields
  • Expert-level knowledge in Ethernet/IP technologies, network topologies, and networking features and protocols in data center
  • Excellent understanding of large-scale network behavior, network simulation, and the effect of distributed computing workloads on the network
  • Demonstrated ability to conduct deep research, prototype ideas, show their value on the path to new technologies, and produce innovative ideas and practical applications
  • Clear verbal and written communication with the ability to build consensus within a large organization
  • Possess problem-solving and critical thinking skills
  • Ability to operate in a highly dynamic environment

Nice to have

  • Practical research and innovations that relate to networking
  • Knowledge in various Switch ASIC hardware and software abstraction
  • Applying ML/AI methods to solve networking problems

What the JD emphasized

  • 12+ years of experience in system architecture, networking technologies, and embedded software, including 3+ years functioning as a Researcher in that fields.
  • Expert-level knowledge in Ethernet/IP technologies, network topologies, and networking features and protocols in data center.
  • Applying ML/AI methods to solve networking problems.