How Data Scientists and Engineers work in tandem to achieve real-time personalization at Overstock
Personalizing online experiences for users is nothing new, but real-time personalization requires sub-second speed and close collaboration between data scientists and enterprise engineers.
Like the hands on a clock, data scientists and enterprise engineers have shifted their focus from hour- hand quickness to minute-hand speeds with a craving to take advantage of each tick of the second hand and personalize in real-time. Previously, daily activities were consumed on improving customers’ experiences tomorrow. Workflows ran overnight when on perm resources were not being tasked. The focus was on the-day-before jobs, always inching forward 24-hours behind.
Since then, we have shifted to hourly jobs and even to tasks that run every five minutes. Finally, we have been personalizing user experiences within the same day and even during the same session. But could we personalize these experiences instantly, immediately, and in real-time? What would that require? What does it look like? Michael Finger and Chris Robinson explore how data scientists and engineers are working in tandem to achieve real-time personalization at Overstock.com
Session hashtag: #EntSAIS10
Michael Finger serves as Overstock’s Director of Software Development in Marketing where he manages the strategic day-to- day operations of email, promotions, and online marketing/AdTech developmentteams. In this role, he has implemented ad biddings, browser push, and affiliate integration. Michael joined Overstock.com in 2007 as a senior software developer, and has since held several roles within the company. He was a crucial member to the software development teams over both productand finance, which integrated bitcoin payment on Overstock.com in 2014. At Overstock, Michael has also managed operations for internal search, personalization, and for Club O loyalty program development teams. He received a bachelor’s degree in Computer Sciences from the University of Texas at Austin and currently resides in Salt Lake City, Utah. Michael volunteers with the Salt Lake County Sheriff’s Search and Rescue Team as an Assistant Commander.
Chris joined Databricks in July of 2019 as a Solutions Architect. Previously, as Director of Data Science for Digital Marketing and Fraud Prevention at Overstock, he and his team utilized big data and machine learning to create personalized shopping experiences for customers. With a focus on working to unite data flow and Data Science to facilitate an omni-channel marketing experience driven by real-time data, he worked extensively on display, email, PLA, and Paid Keywords leveraging modeling techniques from classical time series to cutting-edge Artificial Intelligence. Prior to joining Overstock in 2016, Chris gained widespread experience at early-stage startups using Apache Spark and building out data science frameworks and solutions. He graduated from the University of Utah with dual Masters Degrees in Computer Science and Statistics.