Define Once, Evaluate Anywhere: Building Repeatable and Correct Features at Stripe

Feature engineering can produce a never-ending set of gotchas - from bugs in features that aren't defined the same way (or even in the same language) in training versus production to mistakes in recording when a feature or label was actually available that lead to unrealistically predictive models. We will discuss the system we built at Stripe leveraging event-ed data to enable model developers to quickly define (and test!) complex and highly predictive features in a single place in code and make them available for both training and real-time scoring eliminating some of these common classes of feature generation errors.