Aug 2019 – May 2023 · 3 yrs 9 mos

• Worked on a novel Customer Data Platform for Walmart Labs to build, model and serve the 360° view of both online and store customers in real-time.
• A real-time event-driven distributed QTS (Queue-Task-State) catering to the needs of 600M customers of Walmart , it’s subsidiaries ( Sams , Asda , etc.) and internal Walmart client teams.
• Capable of processing multi-million customer events ensuring low read-write-api latencies, high availability and data durability.
• Developed Identity Graph Conflation, an asynchronous distributed algorithm to co-locate all customer data thereby removing any tail latencies in QTS.
• As one of the founding members of the team, leading the key architectural fundamentals.

Oct 2017 – Aug 2019 · 1 yr 10 mos

• Working on a realtime Distributed Processing and Database platform based on Kafka streams with key focus being on data colocation with processing.

Jun 2015 – Sep 2017 · 2 yrs 3 mos · India

• Worked on Bing’s Knowledge Graph (Satori- Data publishing pipeline)
• Designed a robust feedback mechanism enabling Knowledge Graph to keep good quality data, which improved runtime performance by 2% and reduced storage by 5% on the scale of TBs.
• Sole owner of the pipeline which feeds the data to “People also search for” carousel and in the entity pane for 9 major markets.
• Optimized runtime performance by 80% of a generic ranking algorithm (an adaptation of bag-of-words model).
• Mentor at Microsoft AI Hackathon 2017 at Anna University, Chennai.

May 2014 – Jun 2014 · 1 mo · Hyderabad

• Worked with Bing at Microsoft IDC to build a monitoring, debugging and reporting tool which helps enhancing the Entity Building process of Bing.