Ankit Jhalaria
Godaddy - Principal Software Engineer
Actions
Ever since I was a kid, I was very passionate about computers. When I was in grade IV, I joined a hardware institute in India to build my own PC. Back then, I would use that hand-built PC with a 486 processor to play games like Dave.
My hardware guru introduced me to my first programming language "Visual Basic". I build a very basic application for my dad that would help him keep track of inventory at his tire shop.
After pursing my masters in computer science at USC and interning at CallFire - A VOIP company and Mindjolt - A gaming company, I started working at Yahoo as a Software Engineer in the Ads and Data Platform team. Freshly out of college, I learned a lot about building applications at scale and got introduced to Hadoop and the magic of Map Reduce at Yahoo about a decade ago. After 5 years at Yahoo, I joined a 30 person startup called AtScale to help build a BI platform that can make it easy for customer to query terabytes of data really fast. For the last couple of years, I have been with Godaddy and we are building our streaming data platform using beam and flink to make data available to our downstream customers in a low latency fashion. I am an Apache Beam contributor and love spending time with my son and wife.
Building Stateful Streaming Pipelines
Walk attendees through our journey of building a streaming data platform at godaddy using apache beam and flink. Historically at Godaddy, all of our pipelines were batch and it would take 24-48 hours to make the data available to our down stream consumers. Having such a high latency makes the data not suitable for marketing uses-cases. With a goal of having a low latency pipeline to make data available to our marketing team in a real time fashion, we started writing our streaming data platform from scratch about 2 years ago using beam and have been successfully able to run it in aws using flink on EKS. Having made data available to our care teams [call center] and marketing teams in a low latency fashion has made our customers very happy. We would love to share our journey and learnings during this process of building an end to end stateful pipeline using beam and deploying on flink.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top