Jul 2019 – Nov 2019 · 4 mos · San Francisco Bay Area

Dec 2016 – Jun 2019 · 2 yrs 6 mos · San Francisco Bay Area

• Metamarkets is the leading provider of interactive analytics for programmatic marketing.
• https://techcrunch.com/2017/11/03/snap-metamarkets/
• Primarily responsible for batch processing services that ingest, process and index around 2 PB of data a week
• Worked on improvement and development of in-house timeline oriented Processing and Queuing Frameworks that enables ETL setup without interacting with the underlying processing framework (Spark/Samza) or datastore (Druid)
• Outlier Detection as an Ops monitoring tool based on system and application metrics generated in real time.
• Ease the operation of Spark jobs on Mesos with features such as Maintenance and Checkpointing.
• Help with maintaining and operating one of the largest druid and kafka clusters in the world

Jun 2016 – Sep 2016 · 3 mos · Greater Los Angeles Area

• Worked on multiple production-level data science and engineering projects:
• Designed and developed a scalable, production-level Spark ETL pipeline, replacing legacy SQL pipeline and improving performance speed by 6 times an also reducing cost by 36 times
• Developed customer churn prediction model, provided actionable insights to decrease churn rates.
• Built recommender system using collaborative filtering on Spark for big-data networking app.
• DS12 Residency program: http://education.datascience.com/

May 2015 – Jul 2015 · 2 mos · Banglore

• Part of Mint Windows Team that developed an Universal App for both Mobile and Desktop for Intuit Mint.

Aug 2014 – Dec 2014 · 4 mos · Hyderabad Area, India

• Worked on the lab deployment framework for virtual labs. Wrote scripts to manage containers and to migrate all the labs from Bitbucket to Github.