CMA Developer Code Test, v2

Goals

The purpose of the test is to demonstrate candidate aptitude in the following areas:

• Ability to solve problems creatively and rigorously.
• Ability to understand APIs and use them to solve problems.
• Proficiency in CMA’s tech stack; Python and Mongo specifically
• Strong grasp of how structured data works and how to manipulate it.
• Comprehension of specialized data, in this case CMA’s artwork data.
• Strong documentation and communication skills.

Coding and Problem Solving

Write a script or utility that performs the operations and functions outlined below.

The solution is by no means intended to be a finished product, but the code should be clear, concise, and well commented.

Solutions must be in Python 3.10+.

• Use CMA’s Open Access API to extract a “highlight” artwork based on criteria you decide (e.g. biggest, smallest, oldest, longest artist first name, etc.) from a CMA exhibition of your choosing (current or past) that has at least 8 CMA-owned works (CMA exhibitions often have many works on loan. The API response will only show CMA-owned works. You may have to look at a few exhibitions to find one with more than 8 works!).
• Use the API to find 5 artworks (in or out of the exhibition you chose) that are related by one or more criteria in the artwork data to create a “mini exhibition”.
• Set up a data source in MongoDB Atlas. If you don’t have an account, you can sign up for a free one here: https://www.mongodb.com/cloud/atlas/register.
• Use the Atlas API to insert the following data into a Mongo collection:
  • athena id
  • accession number
  • tombstone
  • image
  • any fields used for highlight or similarity criteria
  • an identifier for your mini exhibition

Documentation

Document the steps necessary to set up and run your application. Include any relevant details with regards to additional modules or runtime environment. This is an important step; running the application will be part of the evaluation process.

Also, briefly answer the following questions about the exercise above:

• How did you approach this problem and why you end up with the solution you did?
• How long did it take you to complete?
• How would you change your solution in order to scale it up to a web application where users can select an exhibition, highlight and similarity criteria, and receive results?
• How would you set this up as a process that runs daily, with different results, and posts them to social media platforms via API?

And Finally...

When complete, post your code and responses to GitHub and send us a link to the repo.