May 2019 – Mar 2020 · 10 mos · Bengaluru Area, India

• Location Tracking Platform, Freight Business
• 1. Designed and implemented an insights platform for metrics based comparison of fleet w.r.t competition.
• 2. Re-designed existing data models and code to make read/writes faster, at higher throughput.
• 3. Worked on POCs using location data to predict pricing on unknown routes.

Aug 2017 – May 2019 · 1 yr 9 mos · Bengaluru Area, India

• Competitive Intelligence, Retail Core Technologies, Walmart US e-commerce team
• 1. Built a system to create matches between products via a bundle of algorithms, with varying degrees of accuracy.
• Built highly scalable web crawlers to scrape ~25 million pages daily, to gather competitive intelligence.
• Removed MySQL dependency and moved to stream-based processing (Kafka), and distributed storage (Cassandra).
• Worked on different Big Data technologies like Hive, Hadoop, Spark, and Flink.

Apr 2016 – Jul 2017 · 1 yr 3 mos · Bengaluru Area, India

• 1. Designed and implemented chat-bot framework for solving post-travel customer queries. Current system solved ~300 tickets per day, and assists customers via interface in Mobile Web Site, and native mobile application.
• 2. Designed and implemented IRCTC integration for GoIbibo (caching layer). The system has zero downtime, and can scale on its own.
• 3. Developed a new flavour of goCash called goCashPlus. Aided in integration of feature with goContacts, a new referral scheme in goIbibo. This feature doubled user acquisition rate.

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

• 1. Developed key features of goCash Wallet like expiry mechanism of promotional goCash
• 2. Developed CRM training modules, query panels, internal tools for financial purposes
• 3. Worked on active online marketing of product, segmentation of users, active communication
• 4. Currently redesigning the architecture of goCash: V2 primarily using rest framework, mongo DB.

Jul 2013 – Jan 2015 · 1 yr 6 mos · Bengaluru Area, India

• Project 1 : Contacts App
• Reduced the run time of contact pairing engine by 50% by implementing JNI layer and efficient algorithms
• Implemented score decay algorithm in display engine for activities; heuristics for joining contacts
• Conducted user audits for accuracy, mean-variance estimation, performance and power profiling on device
• Project 2 : Smart Gallery
• Built engine for categorizing medias, wallpaper suggestions, forming memorable stories from device, SNS data
• Built data mining engine for learning and predicting face tags, geo tags and captions for images
• Built engine for generating short video clips of stories with transitions, captions, and soundtrack
• Submitted paper in Samsung Best Paper Award Contest, 2014; filed for patent consideration
• Project 3 : Location Profiler
• Built data mining service to mine phone context, label and predict location, build user profile
• Proposed strategy to identify key stay points, places of interest, improve accuracy by expanding data sources
• Pitched solutions as enhancements of core apps in Samsung’s flagship models (Galaxy S and Note series)
• Language Used: Java, C (secured Advanced level in internal SW test); Platform: Android; Database: SQL

Aug 2012 – Apr 2013 · 8 mos · New Delhi Area, India

• First author in research paper published in IEEE Indicon, 13
• Title: Improved energy detector based spectrum sensing for cognitive radio: An experimental study
• Project for paper: Spectrum Sensing Cognitive Radio (B.Tech Project Thesis, Apr 13)
• Demonstrated experimentally that Improved Energy Detector performs better than
• Conventional Energy Detector, co-operation and fusion improves accuracy; adjudged second best project in department

May 2012 – Jul 2012 · 2 mos · Bangalore

• Document Classification System
• Implemented document classification system using Natural Language Toolkit and parallelization techniques.
• Implemented XML parser, JSON parser, web page parser, wrote map-reduce pseudo codes for algorithms