Professional Experience
Senior Software Engineer (R & D)
LinkedIn, Mountain View, California
Cortex: Audience Engagement
Technology: Java, Hadoop, Spark SQL, Spark ML, Scala
- Developed ML models to perform sentiment analysis based on engagement data
- Designed offline flow for generating audience segments for custom targeting
Seas AdCreative: Distributed Search for Ads
Technology: Java, Hadoop, Pig
- Developed a general-purpose search service leveraging Galene (based on Lucene index)
- Enhanced plumbing to incorporate diffs in a nearline fashion by listening to changes at databus
Ads Core API: Next Gen Validation Framework
Technology: Java, Hadoop, Kafka, Rest.Li
- Architected the next generation pluggable validation framework for Objective Based Advertising
- Validated dynamic pricing by tuning bid-budget using supervised learning models for autobid campaigns
- Built a Source of Truth by analyzing current state and constraints to compute valid next state
Recognition
- Winner of LMS Craftsmanship Kudos Award Q4'20 - Q1'21 (out of 500+ Engineers)
- Recipient of Bravo Awards in 2017, 2018, 2019, 2020, 2021
- Winner of LinkedIn R&D Ads Contest 2017 for achieving maximum conversions
Senior Software Engineer (Big Data)
Electronic Arts, Redwood City, California
Scoobie: Adhoc Analytics Platform
Technology: Java, Hadoop, Hive
- Designed and developed a multi-tenant querying platform for accessing gameplay data
- Implemented intelligent queuing and query optimization to provide fairness in resource utilization
Intelligent Gameplay Difficulty Tuning
Technology: Java, Hadoop, Hive, Oozie, SparkML
- Developed a supervised learning model to tune gameplay difficulty
- Provided rewards to new gamers to encourage better engagement
Recognition
- Honored to be selected as an Ambassador by Leaders for Good for Human-Centric Leadership (2016)
- Official credit in Star Wars Battlefront (2015) and Battlefield Hardline (2014)
- Awarded Spot Stock Award by EA Digital Platform Data (2014)
- Awarded Customerrific Award for improving Hadoop namenode stability (2013)
Software Development Engineer (Windows Azure)
Microsoft, Redmond, Washington
Analytics Platform for Azure Fabric Fundamentals
Technology: C#.Net, SQL Azure, Windows Azure
- Developed cloud-based framework for accessing deployment statistics
- Created modules for analyzing fundamental aspects (availability, performance) of Windows Azure
RedBox: Running Fabric on a Box
Technology: C#.Net, Managed C++, Windows Azure
- Developed system for running Distributed Kernel of Windows Azure locally
Distributed Search Engine for Azure Logs
Technology: C#.Net, SQL Azure, Windows Azure
- Designed distributed search engine for Windows Azure production logs
- Distributed indexer and merger across clusters to process logs locally
Exempt Non-Officer IT (Proprietary Trading)
Credit Suisse, Center of Excellence, Pune, India
Grid Computing Infrastructure
Technology: SUN Grid Engine 6.1, MPI
- Developed Grid Computing infrastructure for Quant applications
- Set up Grid across HP Blades (running SLES9 OS) and Sun Ultrasparc T2000 processors
- Installed MPI libraries and tested parallel codes on the Grid
Research Experience
Research Assistant
University of Illinois at Urbana-Champaign
Advisor: Prof. William D. Gropp
- Achieved 20X performance improvement of a Parallel (MPI-PETSc) Multigrid Solver
- Developed performance model for Dendro (parallel finite element method package)
- Improved parallel performance (speedup and scalability) of elastic solver
- Made Dendro compatible with latest PETSc from Argonne National Lab