pandas: groupby two columns and return diff for string column

Question

I have a dataframe that looks like the following:

import pandas as pd

data = {\'First\':  [\'First value\',\'Third value\',\'Second value\',\'First value\',\