There is some Python out there, commonly as rwinston says to prototype things, and used in production when a bit slower doesn't matter much.
But as Andy says, the scripting languages are often used as data pumps, sucking data from one system to another, changing the form as necessary.
Because one is dealing with multiple systems which change relative to each other, and because there's lots of low level business logic, scripting languages like Perl, Python, Ruby and even sed/grep and awk are better choices than C++ or VBA.
This is a large % of the real world work on a trading desk, and I'd expect most people around here finding themselves doing it for some time in their career.
A bit of skill in this is useful, and can make you more attractive to employ.
But Andy has it right, that you can get sucked into being a data monkey.
As a headhunter, I willshare with you that some people get stuck doing this for long enough that it hurts their career. A common path is to do this job well, and since data loading is a critical part of the business, they won't let you do anything else.
Also it is not the sort of work where you get much "glory", but it is very easy for a screwup to make you look bad.
The degenerate case is that they decide you are really doing an IT role and move you into the IT department and bonus pool. For confidentiality reasons I cannot give exact details of the path, but this sort of thing has happened to a good number of people.
