Apr 2024 – Dec 2024 · 8 mos · Bengaluru, Karnataka, India · On-site

• Part of Google Security Operations which offers a unified experience across SIEM, SOAR, and threat intelligence to drive better detection, investigation, and response. It collects security telemetry data, applies threat intel to identify high priority threats, drives response with playbook automation, case management, and collaboration.
• I lead the Data management team of Google SecOps. This is a very critical team which is responsible for the intake and normalization of all security related telemetry/logs into the Secops product. The team works on solutions which enables customers ingest large volumes of data (petabyte scale) for threat detection in a self-service, optimized and reliable manner. We are a very customer focused team since ingestion is the first point of contact for the customers. The team architects highly scalable distributed data systems to enable ingesting data at such scale in a SaaS solution

Jun 2021 – Mar 2024 · 2 yrs 9 mos · Bengaluru, Karnataka, India

• Lead a team of passionate engineers responsible for increasing the reliability of Marketplace platform and Uber services in general. The team is responsible for:
• Tier 0 service is a Cache aside (with cache-warmer/refresher) distributed cache implementation in the Marketplace ecosystem which caches Rider, Driver, Vehicle and other entities from Uber’s Core Services, thus decoupling Marketplace’s availability from that of those services. It works as a Latency, Availability and Capacity Cache for the remote services This is a high qps service (peak of 1M rps) very essential to maintain the SLA for core services at Uber and shield them from availability/latency issues from lower tier downstream services.
• We built a platform which aims to quickly mitigate large scale outages by finding the root cause with a few clicks and within a few seconds. It uses Jaeger traces to build call-tree of an API and decorate nodes of call-tree with multiple decorators. Decorated nodes are used to calculate health score , correlate logs, find bad diffs etc. It is a single click solution meant to debug outages/alerts by providing all information in a single UI. It has been very effective and one of the most widely used tools at Uber.
• We built a platform to analyze fail-open/fail-close dependencies among thousands of Uber microservices to identify violations and help correct them.
• Helped build a very simplified developer testing tool called SLATE which has Uber move on from legacy testing tools and save > $1m.
• I also manage a team of contractors responsible for Devops and testing automation.

Jan 2021 – Jun 2021 · 5 mos

Jul 2019 – Jan 2021 · 1 yr 6 mos

• . Moved to Bangalore and took over the management role to grow a team here.
• Grew the Microservices Platform team from 2 engineers to 10 engineers in this short time.
• The team works on developing the microservices platform based on Kubernetes used by various
• services within Nutanix and also customer services on-prem.
• Still a very hands on programmer despite the people management role.

Nov 2014 – Jun 2019 · 4 yrs 7 mos

• Bootstrapped the Microservices platform team. Helped deliver a platform based on Kubernetes for
• running internal services such as Object Store services based on the Nutanix Storage stack. Code
• development was predominantly done in Golang.
• This platform has quickly become the default platform for all new services due to its rich feature set
• like EFK stack for logging, Prometheus for monitoring, CSI driver for integration with Nutanix storage
• stack, etc. The service owners don't need to bother about all this and just concentrate on their business
• logic.
• Worked on migrating the old platform running core Nutanix services to the new Microservice based platform to make it more agile.
• Worked in the Infrastructure team prior to that and helped deliver multiple features such as Self
• Encrypting Drive support, HyperV support, etc. In particular lead the support of HyperV 2016 along with
• one click upgrade from 2012 to 2016 without any downtime.

Aug 2013 – Oct 2014 · 1 yr 2 mos · San Jose, California, United States

• Part of the platform team responsible for the Control path system stack of Skyera OS.

Mar 2010 – Aug 2013 · 3 yrs 5 mos · San Jose, California, United States

• Was part of the team working on Nexus 5000/6000 line of switches used in Data Center Networking. Led the Multi-hop FCoE (Fibre Channel Over Ethernet) implementation. Was also the tech lead for SPAN/ERSPAN/ACL based SPAN. I was the lead on couple of scaling testbeds which included maximum supported configurations for all features. It involved debugging a lot of challenging issues that are a result of interactions of different features.

Jul 2003 – Mar 2010 · 6 yrs 8 mos · Calabasas, California, United States

• Part of the IxOS platform team involved in the development and maintenance of the
• multi-threaded software module for Ethernet cards which directly controls the Ixia hardware Load
• Modules and the CPU’s on each port of the Load Module running the Linux operating system. I did the
• bringup of the LM1000STXS4 load module. The development was done using C++.
• I later joined the Linux Platform team and led the implementation of FCoE/FIP and DCBX features. I was the ead for design/implementation of these features and also interaction with various customers to showcase these features. We ended up onboarding several new customers because of the successful implementation of these features.

Jan 2000 – Jan 2001 · 1 yr · Bengaluru, Karnataka, India

• Part of Consumer Electronics working on Philips DVD Videorecorder.