Solid-liquid interfaces have notoriously haphazard atomic environments. While essentially amorphous, the liquid has short-range order and heterogeneous dynamics. The crystal, albeit ordered, contains a plethora of defects ranging from adatoms to dislocation-created spiral steps. All these elements are of paramount importance in the crystal growth process, which makes the crystallization kinetics challenging to describe concisely in a single framework. In this seminar I will introduce a novel data-driven approach to systematically detect, encode, and classify all atomic-scale crystallization mechanisms described above. I will also show how this approach naturally leads to a predictive kinetic model of crystallization that takes into account the entire zoo of microstructural elements present at solid-liquid interfaces. In this innovative application of data science to materials Machine Learning is employed as an aid to augment human intuition, rather than a substitute thereof. The result is an approach that blends prevailing scientific methods with data-science tools to produce physically-consistent models and conceptual knowledge.