Software Engineer, Network Services

Weights & Biases Weights & Biases · Data AI · Livingston, NJ +2 · Technology

Software Engineer to lead architecture, scaling, and operations of network services for GPU cloud services, focusing on high performance and reliability for AI workloads.

What you'd actually do

  1. Develop, and maintain the network services infrastructure for our GPU cloud services, including networking cloud services, host networking, smart NICs, and overlay/underlay networks.
  2. Collaborate with cross-functional teams, including DevOps, System Administrators, and Cloud Services teams, to ensure seamless integration of network datapath solutions.
  3. Optimize host network performance, latency, and throughput to meet the demanding requirements of GPU-based workloads.
  4. Implement and maintain network security protocols and best practices to protect customer data and infrastructure.
  5. Participate in an on-call rotation, troubleshoot and resolve network-related issues, and provide timely support to the operations team.

Skills

Required

  • network protocols
  • routing
  • security
  • TCP/IP
  • BGP
  • kernel networking
  • Golang
  • Python
  • C++
  • software-defined networking (SDN)
  • network virtualization technologies
  • k8s networking
  • k8s infrastructure
  • CNIs

Nice to have

  • k8s operators
  • k8s controllers

What the JD emphasized

  • highest levels of service quality, scalability and performance
  • demanding requirements of GPU-based workloads
  • network security protocols and best practices