After groupby, how to flatten column headers?

You’re looking for .reset_index().

In [11]: df = pd.DataFrame([[2, 3], [5, 6]], pd.Index([1, 4], name="A"), columns=["B", "C"])

In [12]: df
Out[12]:
   B  C
A
1  2  3
4  5  6

In [13]: df.reset_index()
Out[13]:
   A  B  C
0  1  2  3
1  4  5  6

Note: That you can avoid this step by using as_index=False when doing the groupby.

step1 = step3.groupby(['Id', 'interestingtabsplittest2__grp'], as_index=False)['applications'].sum()