You're no novice to data science, but let's make sure we agree on the basics.
As Peter from DrivenData explained in the video, you're going to be working with school district budget data. This data can be classified in many ways according to certain labels, e.g.
Function: Career & Academic Counseling, or
Your goal is to develop a model that predicts the probability for each possible label by relying on some correctly labeled examples.
What type of machine learning problem is this?