Google Cloud : SDN Platform Engineering for Cloud networking.The primary goal of my team is to accelerate the release velocity of the Andromeda SDN stack within the Cloud Networking domain. Concurrently, we aim to bolster the reliability of pre-production environments. These objectives will be realized through a strategy of shifting left in the Continuous Integration (CI) pipeline and implementing proactive measures to avert incidents in the production environment.𝗞𝗲𝘆 𝗜𝗻𝗶𝘁𝗶𝗮𝘁𝗶𝘃𝗲𝘀• 𝗥𝗲𝗹𝗲𝗮𝘀𝗲 𝗩𝗲𝗹𝗼𝗰𝗶𝘁𝘆: Track Submit2x metrics to identify CI pipeline bottlenecks and optimize frameworks, tools, and test coverage to improve release velocity.• 𝗘𝗻𝗱-𝘁𝗼-𝗲𝗻𝗱-𝘁𝗲𝘀𝘁𝗶𝗻𝗴 𝗙𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸: Maintain a framework for developers to conduct end-to-end testing in a hermetic environment, ensuring high-quality code submissions.• 𝗟𝗼𝗮𝗱 𝗧𝗲𝘀𝘁: Facilitate load and stress testing in pre-staging environments at scales exceeding production to enhance reliability and SLOs.• 𝗖𝗹𝗼𝘂𝗱 𝗡𝗲𝘁𝘄𝗼𝗿𝗸𝗶𝗻𝗴 𝗦𝘁𝗮𝗴𝗶𝗻𝗴: Manage CloudNet Staging, executing extensive end-to-end testing to ensure product reliability and prevent production incidents.

* Raising funds for the partnership between SHF and SGPGI, Lucknow to train the next generation of Pediatric Cardio thoracic surgeons in India. There are 240,000 children born every year with Congenital Heart disease and only 10,000 surgeries are conducted annually. The backlog keeps going up every day and causing one of the largest children mortality rates in India due to lack of surgeons and hospitals with this specialization that are available easily in rest of the world. Please fund this cause - via yourself, your family , your employer. Reach out and learn more at https://www.saloniheartfoundation.org/* Helping SHF's scholarship program where we fund the entire medical degree for children who get into the MBBS program of an Indian Govt. Medical college.

𝗟𝗲𝗱 𝟮 𝘁𝗲𝗮𝗺𝘀𝗖𝗼𝗺𝗽𝘂𝘁𝗲 𝗚𝗿𝗶𝗱 𝗧𝗲𝗮𝗺• Built an abstraction layer to manage Kubernetes resources for applications, enabling Domino application engineers to implement product features without directly interfacing with the underlying platform.• Added features like Forge (based on buildkit) for secure environment builds for data scientists.• Team wrote and open sourced Golang operators to manage compute clusters like Spark, Ray, and Dask.𝗢𝘃𝗲𝗿𝗮𝗹𝗹 𝗜𝗺𝗽𝗮𝗰𝘁: Account teams were able to onboard new customers with varying ML backend requirements to our platform𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿 𝗣𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝘃𝗶𝘁𝘆 𝗧𝗲𝗮𝗺• Migrated Scala to Bazel, improving developer velocity.• Automated merges across feature/release/main branches.• Created a Minikube alternative, providing reliable, production-like environments for developer testing.• 𝗢𝘃𝗲𝗿𝗮𝗹𝗹 𝗜𝗺𝗽𝗮𝗰𝘁: Increased engineering feature velocity by enabling faster iterations and testing, more reliable green builds, and a better CI.

Area Lead for Development Infrastructure𝗥𝗲𝘀𝗽𝗼𝗻𝘀𝗶𝗯𝗶𝗹𝗶𝘁𝗶𝗲𝘀: My teams are responsible for providing an optimal development experience for our engineers.𝗞𝗲𝘆 𝗣𝗿𝗼𝗷𝗲𝗰𝘁𝘀• 𝗨𝗻𝗶𝗳𝗶𝗲𝗱 𝗕𝘂𝗶𝗹𝗱 𝗦𝘆𝘀𝘁𝗲𝗺: The development infrastructure team implemented a unified Bazel-based build system for server and desktop repositories; mobile development integration is forthcoming.• 𝗖𝗜 𝗦𝘆𝘀𝘁𝗲𝗺: Developed an internal CI system that supports 35,000 builds per day across AWS and on-premise instances.• 𝗣𝘆𝘁𝗵𝗼𝗻 𝟯 𝗠𝗶𝗴𝗿𝗮𝘁𝗶𝗼𝗻: The Python team, under the guidance of Guido van Rossum, initiated the migration of our server code to Python 3.

• Driving the growth of Open Stack based private cloud within GoDaddy development community• Working with the GoDaddy product teams to launch new products in the private cloud.• Bringing in a containerized Open Stack control plane deployment for zero downtime, rolling updates with better automation and improved CD/CI, and in the process eliminating legacy home-grown patches and going with latest OpenStack stable release-based control plane cloud for new products• Building a software team that actively participates in the Open-Source Community such as OpenStack Rally and Octavia projects as well as recently open sourced home grown NetMet for tracking data plane reliability• Designing a consolidated network architecture by moving physical network security zones into logical BGP based secure zones, for further data center and rack consolidation and optimizing costs for future build outs, and enabling host-based firewalls, SD firewalls and SDN solutions

𝗢𝘃𝗲𝗿𝘃𝗶𝗲𝘄Delivered and drove innovation for PayPal Cloud, one of the world's largest OpenStack private cloud deployments. Responsible for Infrastructure as a Service and back-office services, supporting thousands of applications through self-service compute, storage, and network for mission-critical production systems and the developer community.𝗞𝗲𝘆 𝗔𝗰𝗰𝗼𝗺𝗽𝗹𝗶𝘀𝗵𝗺𝗲𝗻𝘁𝘀• 𝗜𝗻𝗰𝗿𝗲𝗮𝘀𝗲𝗱 𝗖𝗹𝗼𝘂𝗱 𝗣𝗿𝗼𝘃𝗶𝘀𝗶𝗼𝗻𝗶𝗻𝗴 𝗥𝗲𝗹𝗶𝗮𝗯𝗶𝗹𝗶𝘁𝘆: Improved reliability from 70% to 95% by enhancing cloud deployments through increased monitoring, control plane redundancies, drift elimination, code consolidation in GitHub, and addressing common failure scenarios in OpenStack control plane modules (Nova, Neutron, Cinder, RabbitMQ, LBMS, Puppet, etc.).• 𝗘𝘀𝘁𝗮𝗯𝗹𝗶𝘀𝗵𝗲𝗱 𝗖𝗹𝗼𝘂𝗱 𝗥𝗲𝗹𝗶𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 (𝗖𝗥𝗘) 𝗧𝗲𝗮𝗺: Built a team focused on deployment and automation.• 𝗜𝗺𝗽𝗹𝗲𝗺𝗲𝗻𝘁𝗲𝗱 𝗠𝗮𝘀𝘁𝗲𝗿𝗹𝗲𝘀𝘀 𝗣𝘂𝗽𝗽𝗲𝘁: Enabled controlled and reliable code rollout.• 𝗜𝗻𝗳𝗿𝗮𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲 𝗮𝘀 𝗖𝗼𝗱𝗲: Managed all configuration in GitHub for streamlined configuration rollouts.• 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗱 𝗢𝗽𝗲𝗻𝗦𝘁𝗮𝗰𝗸 𝗨𝗽𝗴𝗿𝗮𝗱𝗲 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗶𝗼𝗻: Reduced data center upgrade time to just one day.

𝗢𝘃𝗲𝗿𝘃𝗶𝗲𝘄Managed the RPD Infrastructure team, responsible for building the infrastructure for routing protocols across the Junos product portfolio.𝗞𝗲𝘆 𝗔𝗰𝗰𝗼𝗺𝗽𝗹𝗶𝘀𝗵𝗺𝗲𝗻𝘁𝘀• 𝗟𝗲𝗱 𝘁𝗵𝗲 𝗥𝗣𝗗 𝗺𝗼𝗱𝘂𝗹𝗮𝗿𝗶𝘁𝘆 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝘁𝗼 𝗰𝗿𝗲𝗮𝘁𝗲 𝗮 𝗺𝘂𝗹𝘁𝗶-𝘁𝗵𝗿𝗲𝗮𝗱𝗲𝗱 𝗥𝗣𝗗 𝗳𝗼𝗿 𝘁𝗵𝗲 𝗡𝗚 𝗰𝗼𝗻𝘁𝗿𝗼𝗹 𝗽𝗹𝗮𝗻𝗲.• Introduced I/O threads, improving BGP and RSVP session 𝘀𝗰𝗮𝗹𝗶𝗻𝗴 𝗯𝘆 𝟰𝘅.• Built a JUNOS tasking library (kevent), benefiting over 60 JUNOS daemons with fine-grained tasking and other enhancements.• Extended JUNOS to support VXLAN for data center deployments.• Converted RPD to a 𝟲𝟰–𝗯𝗶𝘁 𝗱𝗮𝗲𝗺𝗼𝗻 for future scaling.• Reduced backlog by over 70% and incoming bugs through enhanced unit test automation.• Brought a quality focus to RPD through branch and release lead roles.

Managed control plane for Data Center BG. * built out the control plane team for QFX fabric, TOR and microfabric products.

Managed subscriber services * Transitioned and build a new multicast snooping team in Bangalore for JUNOS Mx Platforms

Relocated to India to build a new group responsible for BGP and MPLS for E-Series platforms. * was responsible for 1st software product release from Bangalore and continuing to lead the maintenance release strategy that allows customers to upgrade.* added responsibilties of managing multicast, IP protocols and in charge of overall routing protocols team.

* Responsible for MPLS forwarding (LFIB) Infrastructure for the 7500, 7200 and GSR routers. * Managed some contractors for sustaining effort, did some program management, worked with customer deployment issues and escalations.

worked on MPLS-LSC (Label Switch Controller) with the Stratacom switches. Built the LSC router control plane software on RXM - Router on a blade of one of the Stratacom ATM switches

Founding member of the Rapid Response team for the 7500 where I worked in areas like CEF in IOS.Was involved in troubleshooting customer escalations and rolling out software fixes for the telcos deploying and building the backbone of the internet.

Joined the recent Lightstream acquisition which itself was the spin off from BBN communications - worked on the control plane software of the worlds 1st ATM switch - Lightstream 2020.

Worked on the CANBANK project on Pro*C - implemented several account opening/closing and such banking applications that did transaction processing in C/UNIX and used SQL to embed our queries to the Oracle database. quite an interesting project and learnt a lot.