Detecting Data Drift at Scale

YLabs is building an open source library called YLabs to enable this as a standardized way of thinking about detecting drift at scale. So what we do is again, you have a model with a lot of features and you use those features probably for training first. Then once these signals are collected, then you can run various tests like payout divergence. And then for more complex data problems, you can just let them into more structured information.