Learn how explicitly modeling dependencies in a machine learning pipeline can vastly reduce its complexity and make it behave like a tower of Legos: easy to change, and hard to break.

At Abnormal Security, we use a data science-based approach to keep our customers safe from the most advanced email attacks. This requires processing huge amounts of data to train machine learning models, build datasets, and otherwise model the typical behavior of the organizations we’re protecting.

Our ML pipeline powers a detection engine that catches the most advanced email attacks. These attacks are not only extremely rare, but also change over time in an adversarial way. Since we require both high precision and high recall, and the cost of any error is severe, it is essential...