This is a lecture that organizes things that I vaguely thought were necessary while working on related tasks that require continuous management of ML models. It is also helpful because it talks about how to approach it practically.