Feb 2022 – Jul 2022 · 5 mos

• I was an architect and technical lead in Mobility Data Intelligence team and solving data engineering problems in Rider, Driver and Marketplace domains. To mention a few,
• 1. I have redesigned the complete data flow(Online & Offline) of Jobs data for every product that Uber had that processed every trip(Eats, Grocery delivery, Rides, U4B)/Order that was placed in Uber products.
• 2. Driver Supply being the core data to report/measure/improve business, redesigned and built systems from scratch to bring in data consistency with one single definition of 'Supply Hours' across Uber Rides and Delivery products and made sure all the use cases are supported for one of the heavily used datasets at Uber.
• 3. I had setup the technical vision for the Driver Data Intelligence team and was leading projects in the domain of Growth, Compliance, Vehicles, Sustainability and Driver experience.

Jul 2018 – Feb 2022 · 3 yrs 7 mos

Mar 2016 – Jun 2018 · 2 yrs 3 mos · San jose

• I was building data for the Apple Search Ads(Ads shown in the App store) which powered various business use cases and also used in reporting portal which was used by App developers/Organization in evaluating their Ad campaigns.

Nov 2013 – Mar 2016 · 2 yrs 4 mos

• Yahoo Knowledge Graph
• Knowledge Graph is a knowledge base used by Yahoo search to enhance semantic-search information gathered from a wide variety of sources. It provides structured and detailed information about various entities like actors, movies, books, mobile apps, drugs/diseases and so on. I was involved in designing and implementing acquisition framework, reconciliation framework and enrichment of data. Data acquisition involves gathering data from different sources and getting into the knowledge graph processing systems. Data reconciliation is about grouping similar/same entities from different sources using machine learning algorithms. Data Enrichment is about denormalizing entities so it becomes self-contained and can be used as it without referencing other connected entities.
• Information Extraction
• Information extraction is about extracting information from web or any semi structured data(html/xml). It can be used as one of the sources of knowledge graph to increase the coverage, accuracy and diversity of knowledge graph. Yahoo search uses information extraction for use cases like direct display of election results, sport standings etc., I was involved in
• 1. Hosting a Hadoop application for offline web extractions for knowledge graph
• 2. Design and implementation of backend scheduling systems to refresh data for online extractions for yahoo search

Nov 2005 – Nov 2013 · 8 yrs

• Yahoo Sports
• Yahoo Sports being one of top sites in US, I designed a system to serve critical data like match, scores etc., with latency of few milliseconds. Involved in back end development for Yahoo Sports and Eurosport. I designed database schema for serving the soccer data. Sports was built on top of databases/PHP/web services.
• SpotM
• SpotM was a social networking project. I was involved in developing functionalities like recommending friends from an index based on similar interests, hobbies, location and so on. I was also part of backend design and implementation. It used an indexing store called Vespa for recommendations and MySQL for user store. It was developed for good scale and it involves pretty complex database design, database replication and sharding of user data for scaling the product.
• Yahoo Videos
• This was my first project at Yahoo. This project was meant for serving different videos use cases. I was involved in backend development primarily in PHP and MySQL. I was handling the videos metadata.

Nov 2004 – Apr 2005 · 5 mos · Hyderabad, Telangana, India