The Daily Grind's card reader was down for part of Tuesday, so some orders in the till log have no payment method recorded at all. The owner wants every order that was not paid in cash, and she is clear about the unknowns: an order with no payment method on record might not have been cash, so it belongs in the list.
You have the till_log DataFrame:
Your task: return every order whose paid_by is not "cash", including the orders where paid_by is null. Keep only the columns order_id, drink, and paid_by.
Assign your answer DataFrame to result. Row order doesn't matter.