I set it up here:
It's just a basic python script that I setup...
It found a bad bug which was occasionally seen in production so I've spend most of this week tracking that down.
The other really annoying thing is that I found a bug in Ant which was allowing the unit tests to be marked as passed when they actually failed.
That was no fun...
I think I should have a fix this weekend.