Sep 2020 – Jun 2023 · 2 yrs 9 mos · Hyderabad, Telangana, India

Sep 2019 – Aug 2020 · 11 mos · Hyderabad Area, India

• Leading the Azure Devops search team, that owns all the search experiences in Azure Devops. It is large scale system supporting search 100's TB of data in performant manner. Currently we are working on making our platform more robust and scalable.

Dec 2015 – Sep 2019 · 3 yrs 9 mos · Hyderabad Area, India

• Azure Devops Search - Added support for workitem search and package search in Azure Devops. Users for workitem search are now crossing 500k users. I was security and performance champ for team during this time, and made our platform secure by adding network encryption, authentication and secret management and automated rotation etc. All this via custom solution on top of Elasticsearch stack.
• Also for performance we increased our APDEX score from 0.5. to 0.9 which is industry standard now. This involved improvements at all the layers of product across multiple teams.

Dec 2014 – Dec 2015 · 1 yr · Hyderabad Area, India

• VMWare VM Backups - Added support for backups of VMWare environments, which includes doing prototypes to come up with approaches to take backups of non-Microsoft workloads and then come up with least intrusive design to support that. Had to overcome lot of assumption regarding Microsoft platforms in existing design and properly separate the components so that DPM can work with any underlying platform and not only Microsoft Windows.

Mar 2013 – Nov 2014 · 1 yr 8 mos · Hyderabad Area, India

• 4. Reliability - With large scale deployment we need to consistently meet SLA's. As part of this effort we needed to talk to customer and understand their Environments and needs and make sure that We find the reliability issues early be it in underlying platform, or product and get them fixed.
• Skills - HyperV, clustering
• 5 Cloud Platform Solution (CPS) - CPS is private cloud solution from Microsoft including backups using DPM. This is large scale Deployment and to support this we make Major design changes in VM Backup Stack, including redesigning our File System Filter to support faster storage like SSD and removing dependencies on some platform components which causes issue at large scale.
• Skills - Kernel Driver, Distributed File System (CSV).

Mar 2011 – Mar 2013 · 2 yrs · Hyderabad Area, India

• Data Protection Manager(DPM) is backup product part of System Center Suite. DPM provides backup capabilities for all the microsoft workloads like Fileshare, SQL, Hyper-V, Sharepoint, Exchange etc. Within DPM I am the owner for HyperV VMs and File/Folder backups. Along with this i was the owner of some other core components like Data Source Mover used to move data from production server to backup server and File system filter, a kernel component to track the changes on volume between 2 backups to allow us to do only incremental data transfer during backup.
• During multiple releases of DPM I have owned and worked on following features
• 1. Optimized backups for Deduped Volumes - Deduplication is way to compress data on volume, and I added support in DPM to take optimized backups of such volume i.e. backup the data in compressed form itself saving resources like network, disk etc.
• 2. Adding support for REFS - This is a new File System released by Microsoft and I have added support for backing up this new file system. This included understanding the new File system and making required changes to DPM.
• 3. DPM Scale Out - Since Virtualization is growing the need to backup large scale HyperV deployment grows. To support that we added a new feature to DPM to allow multiple DPM servers to backup single large clusters. This required some large scale changes to make that multiple DPM servers so not interfere each other and do not affect backups of each other. We also have to make some performance changes so as to take backup at this large scale and still meet the Backup SLA's.

Mar 2009 – Mar 2011 · 2 yrs · Hyderabad Area, India

Aug 2007 – Mar 2009 · 1 yr 7 mos · Hyderabad Area, India

• SQLCE is light-weight database (footprint < 2MB) targeted both for device, desktop applications and recently we added support for web server applications. The product offers a rich query processor and support both native + managed programming models. I was the owner of its ADO.Net programming Model, Visual Studio tools for SQLCE, and also worked on its query processor.
• I have also enabled SQLCE to run under .Net Partial Trust environments to be used by Windows Phone to provide its Data API's, and also have patent out of this work i.e. "Securing mixed mode application under Partial Trust"
• Few other project that i worked on
• 1. .Net Embedded database - The goal was to create the whole database engine completely in .Net supporting database size up to 16 TB. As part of that project I was working on prototype, design and implementation of its query processor from scratch along with its programming layer which was based on LINQ.
• 2. Data migration Wizard - Enabling migration of data from in and out of SQLCE databases.
• 3. SQL Azure online management services - This is management portal for SQL Azure service, and I specifically worked on Query plan visualization. Since this was web based application we have to write whole visualization logic from scratch
• 4. Engineering improvements - Helped the team in upgrading our build and test infra along with introducing engineering improvements like gated code checkins, build automation etc.
• Specialties: C#, Database engine, C, C++