Python or SQL solution- Creating effective date and expiration date table

I have a table that tracks bank account information. Each bank account can have someone added as a interested party and taken off at any time (even same day). The person can be added once and never removed or added and removed once or multiple times (even several times a day)

Example:

As seen above person 0015 was added as an Interested Party to account 0004 which is housed at bank 0001. I want to get the data as more of a effective date- expiration date format.

So instead we would see

My data is over 250 million rows, so I am looking for an efficient way to do this. SQL has been extremely slow, so I am trying python. But the only way I can think to do this is to loop through each unique instance of bank/account/person and for each addition find the closest deletion (if any) that is greater than the date of the addition.

Loops are never effective with pandas, so can anyone assist with a clearer more efficient way to gather this data? It could be in SQL too, but this data will never be a new table; it is being used for analytics purposes.

I'm supposing that the data is sorted by the Date. Then you can do:

df["tmp"] = df["Type"].eq("Addition")
df["tmp"] = df.groupby(["BankId", "AccountID", "PersonId"])["tmp"].cumsum()
out = (
    df.pivot(
        index=["BankId", "AccountID", "PersonId", "tmp"], columns="Type", values="Date"
    )
    .reset_index()
    .rename_axis(None, axis=1)
    .drop(columns="tmp")
)

print(out)

Prints:

   BankId  AccountID  PersonId   Addition  Deletion
0       1          4        15   2/9/2022  3/2/2022
1       4          5        38  2/10/2022       NaN