Jul 2024 – Feb 2026 · 1 yr 7 mos · Bengaluru · On-site

• Personalization Data Platform (PDP) provides a one-stop solution for end-to-end personalization data journeys from collection to processing to serving with best-in-class data protection capabilities.
• Individual Contribution
• ∗ Laelaps Integration in Dataflow - Designed the integration of Laelaps as a sink in Dataflow, projecting a 1.8ms latency improvement in Maps path and enabling $XXX Million dollar savings.
• ∗ Koana Column Migration Led the successful migration of several complex columns to Kansas, the new unified personalization data store for inference and training.
• · Reduced Training-Serving Skew Reduced training-serving skew by unifying training and serving
• activity storage and processing.
• · Improved Data Freshness Improved data freshness from 3-6 hours to under 1 minute.
• · Storage Savings Achieved storage savings of 130PB of HDD by migrating to Koana’s more efficient
• storage.
• · Recognition Received a spot bonus from the director for the successful and impactful column migration.
• Leadership & Management
• ∗ Team Building - Built the Personalization Data India team from the ground up, hiring and mentoring a team of 4 Software Engineers to create a nuclear team of 5.
• ∗ Fostering Culture - Fostered a collaborative and high-performing team culture through regular 1:1s, actionable feedback, and public recognition of contributions.
• ∗ Project Execution - Drove project execution and OKR tracking by implementing daily stand-ups and Kanban dashboard.
• ∗ Code Quality - Improved team velocity and code quality by prioritizing timely CL reviews and enforcing best practices.

May 2023 – Jul 2024 · 1 yr 2 mos · Bangalore Urban, Karnataka, India

• I contributed to the development of SpanFS, Cohesity’s distributed file system.
• Optimizing parallelism during reads/writes. Features like in-line deduplication, chunk grouping, and compression enhance efficiency. I played a role in creating the Hydra/Data Journal for caching. I also contributed to sub-services Madrox for replication.

Jan 2022 – Oct 2022 · 9 mos · London, England, United Kingdom · On-site

• Collaboratively developed infrastructure and central visualization tools to enable continuous
• tracking of compliance with committed SLOs and monitor progress. Automated incident escalation procedures based on SLOs and error budget principles.

Jan 2020 – Dec 2021 · 1 yr 11 mos · Hyderabad Area, India

• One Drive and SharePoint (ODSP) | Multi-Geo Team
• With Microsoft 365 Multi-Geo, your organization can expand its Microsoft 365 presence to multiple geographic regions and/or countries within your existing tenant.
• You can provision and store data at rest in the geo locations that you've chosen to meet data residency requirements.
• Users get a seamless experience when using Microsoft 365 services, including Office applications, OneDrive, and Search.

Jul 2019 – Jan 2020 · 6 mos · Hyderabad Area, India

• Master Data Management - Deliver Team
• Implementated Scalable web API for fetching Master data from SQL database.
• o All internal services were using this API for Master Data of any Product.
• o This service needed high availability because it contains critical data.
• o This API removed rudimentary methods of relying on visual inspection of the product, inconsistent inputs gathered over search engines and email communications.
• Role based access to data for Content Manufacturers and Distribution Centers.
• o Appropriate data should be retrieved for specific CMs or DCs, there should not be any data compromisation.

Jul 2018 – Jul 2019 · 1 yr · Hyderabad Area, India

• PowerApps Analytics Team
• IMPLEMENTED FEATURE-GATING WITH CACHING
• This helped us to rollout a new feature to specific tenant/user first.
• This feature was hit by millions of requests, to handle this we have used
• caching.
• CO-DEVELOPED JOB RECOVERY FRAMEWORK
• This framework will re-run the failed Jobs after processing the logs
• provided by the Jobs.
• We have developed a framework on which any service can onboard
• which runs on schedule basis.
• This framework was used to recover terabytes of analytics data.
• IMPLEMENTED SQL TABLE PURGE JOB FOR GDPR COMPLIANCE.
• We could not save user’s personal details in SQL for long time due to
• GDPR violation.
• This job will purge the user’s PII data or stale data after specific time
• period in distributed environment.
• WRITTEN KUSTO FUNCTIONS FOR PROVIDING ANALYTICS OF APPS
• Analytics includes number of unique users, number of app launches in an
• environment.
• These functions are consumed in Microsoft Flows which are triggered
• on respective intervals.It helps us aggregating the data at one place.