Jun 2017 – Oct 2020 · 3 yrs 4 mos · Bengaluru Area, India

• At Jumbotail, I head the technology team for Customer and New Retail Platform.
• We are solving challenging problems in New Retail segment through our J24 Stores by building technologies to manage inventory, purchase and operation. Our integrations provide omni-commerce capabilities to J24 stores, giving the right experience to every consumer.
• I also head the team for our Jumbotail app. We are continuously innovating to provide the best online experience for our B2B customers using technology and data science. Our technology capabilities include merchandising, offers, product, catalogue, search, cart, delivery tracking, schemes and many more.
• Before this project, I built the Logistics System for last-mile delivery, reverse logistics and hub operations. I also developed the Promise engine that uses 14 signals like supply chain capacities, inventory location and service levels to compute right delivery promise in real-time. This engine plays a crucial role in best serving our customers and reducing supply chain costs.

Jul 2016 – Jun 2017 · 11 mos · Bengaluru Area, India

Apr 2015 – Jun 2016 · 1 yr 2 mos · Bengaluru Area, India

• With Shoplane, users can search for products in their city and plan their next shopping destinations with friends. Soon we will launch with apparels and fashion products. The app will also feature a social network, to engage users and help them discover their new look.
• Shoplane also provides a mobile platform for brands and shops to engage with their customers and list their products.
• Our technology stack includes Reactive Systems and search engine built on Luence. Scala is out primary programming language.

Aug 2013 – Feb 2015 · 1 yr 6 mos · Bengaluru Area, India

• ReaderDeck was a search engine, designed to eliminate the need for typing query. The idea was to stop searching and instead start exploring through online resources.
• As part of this project I build a distributed search engine using Lucene, Akka and Spark. Spark was primarily used to process documents and create lucene indexes. Implemented my own ranking algorithm based on the knowledge content of the document, on top of lucene. The distributed search engine was implemented using Akka Cluster. Also worked on creating a knowledge graph from wikipedia and concepts extracted from web document using Neo4J. Everything was developed in Scala.
• The web interface was developed using Chaplin.js (MVC) supported by NodeJS REST services. All backed services used Finagle Thrift.
• This project is currently on hold until I perfect the algorithm further and implement a true knowledge search engine.

Jun 2011 – Feb 2013 · 1 yr 8 mos · Amazon.com Hyderabad

• Worked on various services involved in managing transportation requests and planning/allocating truck and load movement between warehouses. Also created web tools for internal operations team.
• Languages - Java, Perl
• Framework - Spring, SOA (Amazon's internal implementation)

May 2010 – Oct 2010 · 5 mos · Hyderabad Area, India

• Stress testing tool which re created realistic api requests for a SOA service using logs, and produce the difference in response.
• Storing logs on Amazon S3 bucket and with some queryable parameters to fetch them easily. Also implemented og search using Lucene