About me

I am a network architect based in Pune, India, with more than fifteen years of experience designing, building and running data center, hybrid cloud and enterprise networks. Today I work at NICE as a Senior Specialist Cloud Network Engineer, where I design the hybrid network between AWS and on-premises data centers built on Cisco ACI and build the cloud side as code with Terraform.

Before NICE I spent almost six years at VOIS, Vodafone’s shared-services company. I started there rolling out Cisco DNA Center and SD-Access, then led the architecture, deployment and operation of Cisco ACI fabrics for four of Vodafone’s core data centers in the UK and Europe – including the hybrid design with VMware NSX-T, HPE networking and the public clouds, and the Ansible and Jenkins pipeline we used to push fabric changes safely.

My career started in 2011 on an L1 support desk in Mumbai. From there I moved through system administration, team lead, L2 and L3 engineering roles in Nashik – migrating a global WAN from Cisco to Juniper across five countries and rolling out Juniper SRX and EX networks across five Indian cities. That path taught me how networks fail in real life, which shapes how I design them.

I hold the Cisco CCIE Enterprise Infrastructure (#57252) – earned as CCIE Routing & Switching – and have passed the CCIE Data Center written exam. I am also CCNP Data Center and Cisco Data Center Design certified, and certified on Azure networking (AZ-700), VMware NSX-T, Cisco VPN security and Juniper. I have a Bachelor’s degree in Computer Engineering from the University of Mumbai.

At a glance

How I work

• Design for failure first. Every design I sign off has a documented failure scenario, a tested failover and a rollback plan.
• Make trade-offs explicit. Each design shows the options considered, with their cost, risk and operational load, so stakeholders decide with their eyes open.
• Segment by default. Tenants, VRFs, contracts and firewalls are part of the design, not an add-on after the audit.
• Automate the boring, review the risky. Templates and pipelines for repeatable changes; human review for anything that can take down production.
• Troubleshoot with evidence. Show commands, captures and timelines before theories – then a root-cause analysis that prevents a repeat.
• Write it down. HLDs, LLDs, MOPs and runbooks that the next engineer can follow at 3 a.m.

Beyond the day job

I enjoy mentoring junior engineers on change discipline, firewalls and troubleshooting, and I use this site to write about networking, hybrid cloud and how AI can help network teams without putting production at risk.