Oct 2013 – Nov 2018 · 5 yrs 1 mo · San Francisco Bay Area

• As a member of core backend team I worked on building, maintaining and optimizing various of our internal serving and data systems.
• Work summary :
• Scaling and optimizing Blackbird (HBaseCon 2014,2015, 2016) a HBase backed profile lookup service which handles > 50 B requests for large sized data and serves 98% of requests within 40ms SLA.
• Providing various end points for the data in Blackbird to democratize the data for various internal clients via Thrift Service, Hive etc
• Supporting and customizing HBase for various other internal use cases in modeling, data storage on the spectrum of online to offline
• Writing various MR / Hive jobs for optimizing / cleaning up / providing other views of the data in our internal systems
• Building optimized features into our real time bidding system without violating our SLAs
• Supporting current Hadoop Infrastructure, Metrics (Open TSDB) infrastructure
• Bring visibility into some of our real time server applications by exploring and advocating use of flame graphs and optimizing the bottle necks found etc.

May 2012 – Jul 2012 · 2 mos · Bengaluru Area, India

• As an intern I tried to get a good understanding of Google's internal software practices and used them to work on building a fully functional prototype backend server supporting the API to do a complete 2 legged OAUTH2 protocol (RFC 6749) to be used as an reference implementation given out to the third-party domain re-sellers whom google wants to support for their clients and also to use the same implementation to support test infrastructure for this particular product's code.

May 2011 – Jul 2011 · 2 mos · Nice Area, France

• As a novice engineer, worked on implementing/benchmarking heuristics clique finding problem in specialized graphs, understanding good software development practices