Network Automation Software Lead

About the role

TensorWave is seeking a Network Automation Software Lead to design and own a zero‑touch provisioning (ZTP) platform that automatically stands up and validates GPU network fabrics. The role reports to a Software Engineering Lead and combines full‑stack software engineering practices with deep networking expertise.

Key responsibilities

• Own the end‑to‑end ZTP pipeline from bare‑metal switch boot to production‑ready deployment with zero human intervention.
• Generate intent‑based configurations from a network source of truth/IPAM using GitOps‑style deployment, including pre/post‑change validation and safe roll‑backs.
• Establish network validation and pre‑deployment testing using snapshot/digital‑twin techniques.
• Build streaming telemetry and metrics/logging pipelines (gNMI/OpenConfig) to monitor fabric health.
• Instrument GPU‑specific metrics such as RoCE health, PFC/ECN counters, optics/link errors, BGP/EVPN state, capacity and utilization.
• Deliver actionable dashboards and alerting for the network team.
• Gather requirements, create self‑service APIs, and accelerate deployment velocity.
• Hire, mentor, and lead a small team of software engineers and SREs, setting roadmap and technical standards.
• Integrate the platform cleanly with broader infrastructure provisioning, CI/CD, secrets, identity, and observability stacks.

Required profile

• 8+ years of relevant experience in software engineering and network automation.
• Proven track record building large‑scale network automation solutions, preferably at a hyperscaler, large cloud provider, or massive datacenter.
• Experience leading small engineering teams and driving technical direction.
• Strong understanding of GPU network fabrics and associated protocols.

Required skills

• Zero‑touch provisioning (ZTP)
• GPU network fabrics
• IP Address Management (IPAM)
• GitOps deployment practices
• gNMI and OpenConfig
• RoCE, PFC, ECN counters
• BGP and EVPN
• CI/CD pipelines
• Source control, testing, code review, release management