Nope, we're just using the standard public API and calling the last 200 tweets. To get at more than 200 we have to use some ugly looping that breaks more often than it works.

http://dev.twitter.com/doc/get/statuses/user_timeline