We started out discussion by asking some questions in the form of queries about the data stored in datastore. Some of the questions were answered and for some, we met with exceptions. Next, we embarked on the journey of how queries actually work under the hood. We learned that all queries are actually just translated into the BigTable row scans and for this, separate BigTable tables are maintained, which are updated whenever you write any new entities or modify or delete any existing ones. We learned the role of indexes and how they are not an optional thing but a required component unlike indexes in the RDBMS world.

Armed with this understanding, we went on to explore the Query API. We learned that all the magic happens around the ndb.Query object with its filter() method. We learned how to query for repeated and structured properties. We next focused on how to perform the logical AND, OR operations, how they work under the hood, and how an OR operation just merges two different...