Sr. Network Engineer

Vellstone; Mumbai; Presencial; INR 4.000.000 - 7.000.000; Jornada completa

Descripción de la vacante

Vellstone is seeking a Senior Network Engineer in Mumbai to design, deploy, and operate hyperscale EVPN‑VXLAN Clos/Leaf‑Spine fabrics and InfiniBand networks for GPU clusters and HPC workloads. You will own the full network lifecycle, from requirements to day-to-day operations, including lossless Ethernet and host‑side Linux networking for AI workloads.

The role emphasizes hands-on management, automation using Python/Ansible/Git, and cross-functional collaboration with vendors to rapidly resolve

Formación

• 7+ years of experience in networking fundamentals and data center architecture, with at least 5 years hands-on with BGP, OSPF, EVPN, and VXLAN.

Responsabilidades

• Design, build, and operate hyperscale EVPN‑VXLAN Clos/Leaf‑Spine fabrics and InfiniBand networks for GPU clusters and HPC workloads.
• Own the complete network lifecycle from requirements gathering to lab validation, deployment, and day-to-day operations.
• Deploy and tune lossless Ethernet (RoCEv2, PFC, ECN, DCQCN) and host-side Linux networking for zero-loss AI training and inference.
• Develop robust network automation pipelines using Python, Ansible, and Git.
• Monitor network health, diagnose production issues at packet level, and coordinate with vendors for rapid resolution.

Conocimientos

BGP

EVPN

VXLAN

OSPF

Leaf-Spine

InfiniBand

RoCEv2

PFC

DCQCN

Python

Ansible

Git

Linux networking

GNS3/EVE-NG

Herramientas

Arista EOS

Cumulus Linux

SONiC

Juniper

Cisco

Nokia SR Linux

Descripción del empleo

Experience: 7+ years (including 5+ years hands‑on with EVPN, BGP, and VXLAN)

We’re hiring a Senior Network Engineer for our client, an AI Cloud company helping enterprises build, scale, and govern AI in production.

As the Senior Network Engineer, your responsibilities would be:

• Design, build, and operate hyperscale EVPN‑VXLAN Clos/Leaf‑Spine Ethernet fabrics and InfiniBand networks for GPU clusters and HPC workloads.
• Own the complete network lifecycle, from requirements gathering and architecture design to lab validation, deployment, and day‑to‑day operations.
• Deploy and tune lossless Ethernet (RoCEv2, PFC, ECN, DCQCN) and host‑side Linux networking to ensure predictable, zero‑loss performance for AI training and inference.
• Develop robust network automation pipelines and Infrastructure‑as‑Code practices using Python, Ansible, and Git.
• Monitor network health, diagnose complex production issues at the packet level, and collaborate with vendors for rapid issue resolution.

Requirements

• 7+ years of experience in networking fundamentals and data center architecture, with at least 5 years of hands‑on experience with BGP, OSPF, EVPN, and VXLAN.
• Strong expertise in Clos/Leaf‑Spine fabric design, multi‑tier network topologies, and capacity planning.
• Hands‑on proficiency configuring, testing, and validating major network platforms (Arista EOS, Cumulus Linux, SONiC, Juniper, Cisco, or Nokia SR Linux) in production.
• Practical knowledge of lossless Ethernet (RoCEv2, PFC, ECN, DCQCN) and InfiniBand interconnects for GPU/AI or HPC workloads.
• Advanced network automation skills using Python, Ansible (or Salt), and Git‑based CI/CD workflows.
• Strong understanding of Linux networking, packet‑level troubleshooting, and topology simulation tools (NVIDIA Air, GNS3, or EVE‑NG).
• Build foundational hyperscale network infrastructure powering production AI clouds for leading enterprises, startups, and research institutions.
• Solve complex, cutting‑edge engineering problems at scale across lossless RDMA fabrics, InfiniBand, and massive GPU clusters.
• Work directly alongside and learn from seasoned industry pioneers who founded NetScaler and Netmagic.
• Enjoy end‑to‑end ownership across architecture, automated deployment, and production operations.
• Be part of an engineering‑first culture focused on high ownership, technical craftsmanship, and continuous learning.