This document is an introduction to extreme gradient boosting with XGBoost. It begins by discussing the basics of supervised classification, decision trees, and boosting. It then provides an overview of XGBoost, explaining that it is an optimized gradient boosting library that achieves state-of-the-art performance for many machine learning tasks. The document demonstrates XGBoost code examples and discusses decision trees, boosting, and when XGBoost should and should not be used.