There are a number of places that I know with specific BlackBerry administration and support groups, but these engineers are typically interfacing with end-users for support, wireless acquisitions, deployment, etc.
The truth of the matter is that BlackBerry server administration is not overly difficult and is not a full-time position for 99% of corporate deployments.
There are some "unique" deployments, such as large insurance firms (StateFarm, as an example), large banks (CitiGroup, Bank of America, Wells Fargo), healthcare, and government sectors, that require full-time BlackBerry server administrators. These environments have tens of thousands of users (or in some cases, hundreds of thousands) across hundreds of servers and it is a challege to have both the skills necessary to manage these environments, in addition to pushing the technology to its limits (many times without vendor experience with or knowledge about these types of large-scale deployments).
For my job, I am with the messaging team, so we oversee Exchange, Enterprise Vault, RightFax, BoxTone, and all mobility platforms. I also assist with the overall Windows server administration goals and tend to be well-versed and very much involved across all of our environments. We have 5,000-plus BlackBerry devices and around 4,000 ActiveSync devices.
The best advice I could give to a BES specialist in my type of environment is to ensure that you have designed your environment to be extremely stable (not withstanding the connection points, such as the messaging environment and the network, of course) and implement monitoring to some degree. The ease of mastering BlackBerry server administration (which doesn't necessarily account for random bugs that pop-up from version to version) is perhaps one of the quickest trades to learn in the entire IT industry. With a stable platform, you can focus your time elsewhere ... and that additional and broader experience can equate to one hell of a personal portfolio.
My environment has had 1 hour of unexpected downtime in the last 15 months. This involved a single messaging agent that simply stopped processing and this was not a condition that our monitoring was configured to detect (even RIM said they had never seen it). This affected was a couple hundred users in late evening hours, so the total impact was minimal at best. Aside from security updates, basic user clean-up (which I should script), and review of incidents/requests generated to my team's queues, there's simply no involvement on my part.
Prior to my arrival at my company, we had over 5,000 tickets relating to BlackBerry sent to us in the prior year (1:3 ratio for requests to incidents). In the year of my arrival, the ratio of requests to incidents reversed (3:1). Last year (three full years into my project), we had 22 tickets sent to us (only 3 of which were incidents). This year, we have had 6 tickets sent to our queue (1 incident). The foundation of how productive you can be as a holistic IT person is directly related to the design and architecture of your environment. Just keep that in mind.