If you would be up for a slightly different structure, you could take advantage of file partitioning and write to an output partition of MM_YY. For example, with a dataset that is set up in this format, the output files will have the structure of MM_YY/out-s0.csv.gz:
However, it is not possible to change the actual filename using this method. Instead you can only set the parent subfolders. If setting the filename itself is important to you, then the main option would be to create the file through a Python recipe, so that you can specify the output filename. You could use he managed folder APIs to write to an output managed folder on S3.
Please let me know if you have any questions about these approaches. It might help to see some screenshots of your flow as well, in order to understand your exact use case.