This document presents an overview of techniques for exploring embeddings of time series data, including state space reconstruction, the embedding theorem, and minimal embeddings. It discusses using embeddings to infer properties of dynamical systems from time series data and determine optimal time delays and embedding dimensions. Methods are presented for measuring coupling and causality between variables, including continuity statistics to determine if a continuous functional map exists between two sets based on rejection of the null hypothesis that points are landing in epsilon balls by random chance. Applications to time series modeling, feature selection, and outlier detection in machine learning are discussed.