ne of the most frequently asked questions I hear is, “How do I get a job in R&D?” First of all, there aren’t actually any openings right now, but I thought it might be interesting to look at how little pieces of the process works when we do need to hire somebody new. I’ll focus on what I think is the most interesting part – the “Vapor Ops” test that is given out to prospective developers.
Whenever any job anywhere at Wizards is opened up for public applications, it gets posted on our website at http://www.wizards.com/default.asp?x=company/jobs. (This includes everything from R&D hiring a new developer to the mail room hiring a new clerk.) Then, of course, we collect résumés and pick out the promising candidates.
One thing that really helps at this stage of applying for a job in R&D is if you’ve made a name for yourself in the Magic community. Each of the five developers that have been hired during the last four years was someone with significant Pro Tour experience. Since one of the most important jobs of a developer is to help test new Magic cards, we know we need candidates that are capable of building good, new, innovative, competitive decks. Well, the Pro Tour is the obvious place to look for people that fit that bill. In many cases (including when I was hired) someone from R&D will notice that a certain PT player seems to have the tools to make a good developer and will take them aside, explain the job that is opening up, and recommend that they apply. For me, it was Mark Rosewater who sat me down to make that recommendation and it was at US Nationals in 1999 (which was at Origins that year). He introduced me to Bill Rose (who was then the Vice President of Magic Development and is now the VP of all of R&D) and a couple of months later Bill decided to hire me. (Of course, Bill’s real claim to fame is that he won the Wizards Invitational two weeks ago.)
Being a “name Pro Tour player” can also help even for jobs outside R&D: When Aaron Forsythe got his job managing this website and when Alan Comer got his job programming Magic Online, both were able to point to that experience as definitive proof that they had the deep understanding of the game that was a requirement of their jobs.
When people ask me what the best thing they can do is if they want to get a job in R&D some day, I usually say to study math or science or engineering, because the ability to think logically and rigorously is really important. However, the most common entry-level position in R&D is Magic Developer and the more I think about it, the more I realize that the best way to get your foot in the door is to do well on the Pro Tour (and get to know the R&D types that always attend).
To be fair, there’s a lot more to working in R&D than just testing the power level of new Magic cards. We also need people who can channel Timmy or Johnny and we need creative people who can think up new ideas. In addition, we work on a lot of other games like Star Wars, MLB Showdown, and Duelmasters. But almost everyone in R&D spends some of their time on Magic so being able to contribute as a deckbuilder is a skill we value highly.
The real fun starts once the résumés have been collected and the promising candidates have been identified. That’s when WotC sends out the battery of tests that will help narrow things down to the handful of candidates who will be interviewed. When I applied I had to take a rules test, I had to submit a writing sample, I had to recommend three cards to reprint and three cards that should never be reprinted again, and I had to give my opinion about all 70 of the cards on the Vapor Ops test.
The name “Vapor Ops” comes from a Netrunner concept. The idea was that the corporation could set up fake operations for the runners to break into and steal data from. The runners would think they had real data, but of course none of it was legitimate.
The Vapor Ops test works pretty much the same way – it’s a bunch of Magic cards, some fake and some real, and the test-taker has to try to figure out what’s going on with each of them. What usually happens is that the Vapor Ops test is a mix of cards that have already been printed, cards that have already been developed but have not yet been printed, and cards that R&D would never dream of printing. The evil test-making corporation also goes in and changes some of the words and numbers, deliberately planting many different kinds of mistakes (both obvious and subtle) just to see how many of them any given candidate will catch. Really, though, there are no “correct” answers to the questions on the Vapor Ops test. How you say something is more important that what you say – that is, the people grading the test are really looking to see if your opinions are clear and well thought-out.
It’s been kind of fun for me to go back at look at the test I submitted back in ’99. I get to compare what I thought then with what I think now, which is particularly fun for the cards that were released in the Masques block (after I took the test). Accumulated Knowledge was on the test (as “Card Flare”) but they had it priced at . I’m proud to say I saw their trick: I argued that price would only make sense if it could see itself when it resolved, but it wasn’t in the graveyard yet, so I recommended they drop it to . I also correctly caught their change to Foil – they had it so you could discard any two cards, but that would mean it could go into any deck so I said one of the discarded cards needed to be an island or a blue card. Of course, I also said that Squee (who was called “Anvilhead” on the test) wasn’t very good and should be dropped to . I was thinking of him as a Brood of Cockroaches variant that could only come back if he went to the graveyard from play – I missed the fact that you could use him to fuel Survival of the Fittest, Masticore, etc.
The real problem with my test answers, though, is that I was obsessed with costing issues. A vast majority of my 70 answers includes comments like “this doesn’t need to cost this much” or “too good in Limited at this price.” However, there’s a lot more to Magic development than just tuning the mana costs of all the cards. It’s crucial to make sure that the cards are fun to play with and interesting to think about. They also need to be understandable and they have to have proper flavor. I think I got a little bit lucky that Bill was specifically looking for someone who could obsess over costings and so my Vapor Ops test was good enough to get me through to the interview stage.
After the tests are submitted there are interviews. There were four people flown in for interviews when I was a candidate, and two of us wound up getting hired (myself and Mike Donais). After that it’s time to learn all the lessons of Magic development (a process that really never ends) and next time you get to be on the other side of the Vapor Ops test.
There you have it – that’s what R&D does when it needs to hire a new developer.
THIS WEEK’S POLL:
Are you interested in hearing more about the Vapor Ops test? (In other words, should I use a future article (or two) to look at actual questions and answers and let you try your hand at them?)
Do you want more Vapor Ops stuff?
LAST WEEK’S POLL:
Not nearly as lopsided as the poll from Goblin Week, but once again the winner is a card with both Spike and Johnny appeal.
Randy may be reached at firstname.lastname@example.org.