PRKL - Python Flask API & MongoDB showcase

Requirements:

Use python with Flask.
Use a MongoDB server for storing the data provided in the file "songs.json".
All routes should return a valid json dictionary.
Write tests for the API.
Follow the KISS principle.
Provide all instructions to do the setup, the easier it is for us to get it running the better.
Please take into consideration that the number of songs and ratings will grow to millions of documents as well as the number of users using the API.

GET /songs
- Returns a list of songs with some details on them
- Add possibility to paginate songs.
GET /songs/avg/difficulty
- Takes an optional parameter "level" to select only songs from a specific level.
- Returns the average difficulty for all songs.
GET /songs/search
- Takes in parameter a 'message' string to search.
- Return a list of songs. The search should take into account song's artist and title. The search should be case insensitive.
POST /songs/rating
- Takes in parameter a "song_id" and a "rating"
- This call adds a rating to the song. Ratings should be between 1 and 5.
GET /songs/avg/rating/<song_id>
- Returns the average, the lowest and the highest rating of the given song id.

Design a high-level system architecture to support the above.
If you’re comfortable with AWS, define the AWS components to support the system architecture design.
Describe what is your strategy for scaling each components of the system. Imagine millions of users and on weekends, the traffic could easily spike to 3-5x the average traffic.
Think about components that may be required to monitor and maintain the system
Describe and justify the tooling you would use for development, continuous integration and running the application on prod.
You can use any diagramming tool you wish (one example could be http://dia-installer.de/). You can also use a pen and paper and take a photo.