If you're neck-deep in the Android scene, you've likely heard the word "automagically". The word comes up in the rooting scene a lot and is used by Google quite often, as well. Essentially, it's a step above "automatically", implying that the thing happening automatically is so advanced, so perfectly done, so fast, so frequently updated, or otherwise so perfect, that it's comparable to magic. That term is what Google is using to describe a new update to Google Keep, rolling out today for web, iOS and Android, that allows users to keep on top of their own notes by searching them based on categories that the app makes for them.
The update is deceptively simple when viewed from the front, but there is a lot going on under the hood. Essentially, Keep analyzes your notes in detail and assigns them categories based on their content. Categories like places, travel and food pop up "automagically" when you hit the search button, and tapping or clicking one will take you to all notes assigned to that category. A traditional search for that category will also yield those results. The categories are fairly broad, meant more as an organization tool than a comprehensive tool for finding a single note among many, but the way it all works is still fairly impressive, despite the deceptively small user-facing change, and leaves the function open for future improvement and iteration, as well as forking to provide functions like presenting notes when Google detects you'll need them, for example.
Getting the topics down like this new tool does is just another fruit of Google's push to put machine learning at the core of everything they do. Usually reserved for high-level AI, Google is experimenting with just how many places they can shove machine learning to enhance users' lives, and having a computer deeply analyze your Keep notes and figure out what they may pertain to for future indexing is just one example of that commitment to experimentation and zealous implementation finding its way to users. Google also uses machine learning in Search, and in their public cloud and infrastructure as a service offerings.