A Unified ML Data Pipeline for Real-Time Features: From Training to Serving
On a global marketplace like Etsy where buyers come to buy unique, varied items from sellers from around the globe, the inventory of items is constantly changing. Users preferences also change in real time as they discover the latest selection being offered. In such a dynamic environment, Machine Learning models for different applications (including search, recommendations or computational advertisement) need to collect different real time data signals, process them and finally leverage them to make the most relevant predictions.
In this talk we will detail how we use realtime feature logging & streaming systems to capture in-session / trending activities, in order to compute features for our different ML models and use it for downstream applications such as a Bandit or Reinforcement Learning System.
Anni is a Senior Software Engineer working on Etsy’s ML systems. Her work ensures that Etsy’s buyer activity is readily processed in real-time and easily integrated into ML applications that powers in-session personalization for search and recommendations across the organization. Anni holds a degree in Systems Design Engineering from University of Waterloo and a masters degree researching ML applications in healthcare. Anni lives in Toronto and in her spare time enjoys venturing into the Canadian wilderness with her canoe.