Tensordyne

Tensordyne

System Software Engineer, Networking

Sunnyvale

Sponsorship not specifiedDetected 7 days ago
PythonGoC++Full-Stack DevelopmentCI/CDLinuxLLMsElectrical EngineeringBGP/OSPFLeadershipCommunicationMentoring

About the role

  • This is a deeply technical, hands-on role for someone who is at home moving up and down the stack - tuning userspace services, patching kernel drivers, and reaching into firmware - to squeeze latency, bandwidth, and reliability out of large, lossless networks.
  • The ideal candidate treats "the network" as one continuous system spanning application, operating system, ASIC, and optics, and is energized by chasing a single packet path across all of those boundaries.
  • This role is based on-site at our Sunnyvale, CA headquarters.

Responsibilities

  • Optimize lossless Ethernet and RDMA/RoCE transport using congestion-control mechanisms such as PFC, ECN, and DCQCN, including buffer and QoS tuning tied to real workload behavior.
  • Drive route, neighbor, next-hop, and ECMP programming flows from control-plane intent all the way to confirmed ASIC hardware state, and build mechanisms that distinguish control-plane acceptance from SDK acceptance from hardware programming acknowledgement.
  • Select and validate switches, optics, transceivers, and cabling systems, and lead fabric bring-up, link training, SerDes tuning, and performance baselining together with hardware and platform teams.
  • Develop automation for provisioning, configuration management, and remediation, and contribute to reproducible build and rollout pipelines for network software and images.
  • Define deployment standards and operational best practices so fabrics can be stood up consistently across sites, and support the path from factory qualification through mass deployment.
  • Partner closely with silicon, hardware, platform, and manufacturing teams, and help evaluate next-generation networking technologies and vendor solutions.

Requirements

  • Strong experience designing and operating data center or HPC networks, ideally supporting distributed AI or other tightly-coupled workloads.
  • Strong hands-on skills in network configuration, debugging, and performance tuning, and a track record of chasing hard problems across software, hardware, and vendor boundaries.
  • Excellent technical communication skills, with the ability to explain complex networking concepts and practical trade-offs clearly.
  • BS or MS in Electrical Engineering, Computer Science, or a related technical field, or equivalent practical experience.

Nice to have

  • Experience with SONiC or comparable network operating system stacks (FBOSS, Cumulus Linux, or equivalent platform software), and with SAI implementations.
  • Experience integrating or debugging switch ASIC SDKs from Broadcom, Marvell, NVIDIA, Intel, or comparable vendors.
  • Platform and board-level bring-up experience touching thermal and fan control, power monitoring, transceivers, watchdogs, CPLDs, OSFP/CMIS, or LEDs.
  • Experience with OpenConfig gNMI, YANG data models, or other structured telemetry.
  • Prior technical leadership or mentoring experience.
  • Additional Information
  • To comply with U.S. export control laws and regulations, candidates for this role may need to meet certain legal status requirements as provided in those laws and regulations.
  • Tensordyne is an equal opportunity employer.

Compensation

  • We take care of our people and their families with comprehensive benefits, competitive compensation, flexible spending options, and recognition programs — because category-defining technology starts with a healthy, supported team.

Benefits

  • Build and maintain high-resolution telemetry pipelines that surface network health, detect microbursts, and expose congestion patterns across the fleet.

Company info

  • What We're Looking For

Visa & Work Authorization

  • export control laws and regulations, candidates for this role may need to meet certain legal status requirements as provided in those laws and regulations

This listing is sourced directly from Tensordyne's careers page and normalized into a canonical job model.