Nov 2020 – Mar 2024 · 3 yrs 4 mos · Cupertino, California, United States

May 2019 – Nov 2020 · 1 yr 6 mos · San Francisco Bay Area

Dec 2017 – May 2019 · 1 yr 5 mos · San Francisco Bay Area

• Analytics Platforms and Applications (Part of Data Org at LinkedIn)
• Work on massive data flow architectures for LinkedIn Marketing Solutions (Linkedin Ads)
• Creating scalable Data Warehousing solutions based on Spark and Presto.
• Part of Proposal Review Committee for Microsoft MLADS(Machine Learning and Data Science) Conference - Fall 18 and Fall 19

Sep 2016 – Dec 2017 · 1 yr 3 mos

• Led and Managed the backend reporting and analytics team of Drawbridge's Ad Platform products.
• Also led the ads targeting team for brief time as a joint responsibility.
• Both hands on development of the data reporting platform to develop near real time capabilities and also managing a team to achieve state of the art reporting capabilities.
• Worked on reporting pipelines/Data pipelines that process more than 100 Billion events per day.
• Product facing enhancements for the data pipeline.
• Drove and engineered a Presto based Data Warehouse, that enabled Business Analysts, BI engineers and Data Scientists to query the datasets on HDFS directly as against a serving layer columnar relational DB thereby increasing the amount people can query and use in their models.
• Analytics dashboards based on Data Warehousing
• Backend Technologies : Hadoop, Hive, Pig, Spark, Hbase, Couchbase, Oozie, Columnar DBs, etc.
• Key Project Contributions :
• 1. Conversion Tracking/Attribution https://drawbridge.com/blog/p/the-drawbridge-approach-to-attribution
• 2. Cross Device Insights with Real time Attribution Metrics https://drawbridge.com/c/insights&attribution

Sep 2016 – Apr 2017 · 7 mos

Dec 2013 – Oct 2014 · 10 mos

Jan 2013 – Jan 2016 · 3 yrs

• @Labs - WMX (Walmart Exchange)
• Data Mining and Analytics for Online Ads
• Ad Impression Ingestion and Processing : As an early stage member on the team implemented and fully owned (dev, test, and product) the Ingestion pipeline processing upto 1B impressions per day from various sources.
• Designed and implemented terabyte scale analytic pipeline
• Worked in a small team to build a lambda-architecture realtime analytics and OLAP platform from the ground up
• Worked with Hadoop/Hive and Apache Spark to process ad impressions data.
• Pioneered use of Apache Spark on the team for improved pipeline performance and maintainability
• Real time impression ingestion with Apache Storm (experimental).
• Real time analytics.
• Built a querying system using Apache Solr to build analytics item sets using Clustering of Short Strings in Large Datasets
• Languages used: Java, Python

Jul 2012 – Dec 2013 · 1 yr 5 mos · Sunnyvale, CA

• Media Foundation - Content
• Development related to Content Enrichment for all ingested content in the Content Agility pipeline, and served up on all of Yahoo Media properties (Yahoo News, Yahoo Sports, Yahoo Finance, etc.)
• End to end Design and Development related to selection of most relevant Canonical URLs and contextual clickthrough URLs for content. These power all Yahoo hosted content urls on all content streams including the Yahoo homepage.
• Sole developer for Canonical URL generation for all content in the Enrichment Workflow to achieve Search Engine Optimization for each content (story, photo, video)
• Independently Developed Content Enrichment Libraries for Categorization, Content Quality marking, Geo Classification, Named Entity Recognition, etc. on content (stories, photos, videos) through Feed Ingestion and Editorial Ingestion Workflows using Yahoo’s Contextual Analysis Platform.
• Enabled the Ads Categorization to be tagged for each content.
• Development within Category Taxonomy Management Workflow for Yahoo Content Taxonomy.
• Setting up of Continuous Integration Environment for Content Track’s Libraries.
• Ramping up towards development related to Cassandra and HBase. (Performance Testing of Hector and Astyanax client and different data storage schemas for low latency).

Aug 2011 – Sep 2012 · 1 yr 1 mo

• Fundamentals of Embedded Systems

Jun 2011 – Aug 2011 · 2 mos · San Francisco Bay Area

• Development of Browser Performance Bench-marking Suite for Android Devices

Jul 2009 – Jul 2010 · 1 yr · Pune/Pimpri-Chinchwad Area

• Worked on the Expert Systems and HealthCheck products.
• Expert System was a rule based engine (rule engine) in Java with the Rete II rule engine. Used the OPSJ rule engine for building the rulesets, rule engine.
• Worked for Avaya's R&D (TS&D) team which develop tools & components for easy maintenance & fault-resolution of Avaya's suite of products.
• http://www1.avaya.com/campaign/demo/expert/