Working (and playing) for Valve

Working (and playing) for Valve

Working (and playing) for Valve

With the increasing prominence of game industry ‘personalities’ – more often than not serving as the mouthpiece for larger studios’ projects – it’s difficult to shake the picture of the auteur issuing orders down a chain of command, creativity restricted very much to the top tier. Of course, many developers’ experience of working on a big-budget, potentially massmarket, title will be one of focused specialism, and necessarily so. But Valve has built an enviable reputation for thinking outside of the box – orange or otherwise – when it comes to game development, and this ethos permeates every single one of the company’s decisions, including those centred on recruitment.

By now it’s well known that in 2005 Valve hired the entire team behind Narbacular Drop, a project created by senior students at the Washington-based DigiPen Institute of Technology. The ideas that formed the core of that 15-minute game eventually became the much-lauded Portal. Far from finding the move from education to working for a large company restrictive, however, the team discovered a surprisingly relaxed approach at Valve.

“I can remember a few things that we were told not to do at DigiPen. We did them anyway, but those conversations never happened at Valve,” explains programmer and DigiPen graduate Jeep Barnett, who recently masterminded The Potato Sack ARG for Portal 2’s release. “They really left us alone to do things the way we wanted.

“Most of our team had already found jobs at GDC and were starting to part ways when Valve invited us to come back together for Portal. It was pretty sudden for Dave [Kircher, now a Valve software engineer] and me since we were nearing completion of our first title at Sandlot Games. We actually worked both jobs for an overlap of three months!”

The success of Portal no doubt contributed to Valve’s decision to repeat the trick of hiring an entire team straight from university, which it did after seeing another DigiPen project, Tag: The Power Of Paint, in 2008. The surface-property altering paints of Tag became a major new mechanic in Portal 2, in the form of Aperture Science’s experimental gels.


Level designer and programmer Tejeev Kohli (left) and programmer Jeep Barnett

“When we were looking to work on Tag after graduation we had thought about starting our own company and getting funding from a publisher,” recalls Portal 2 level designer and programmer Tejeev Kohli. “We didn’t really know much about the business at the time and I’m certain that had we gone through with it, we would have failed miserably!”

Kohli goes on to echo Barnett’s sentiment: “I would say that I have more freedom at Valve than I did when I was at DigiPen. At school I had five different classes a semester and had to do assignments for those, on top of working on the game project. At Valve I’m working on the game full-time and am making the decisions on what I want to work on.”

Both teams were hired on the strength of the projects they’d already completed, an achievement Barnett and Kohli stress is essential for graduates looking to get noticed by established studios today. And, just like Kim Gordon rearranging the basslines of other bands’ songs to create early Sonic Youth material, the easiest jumping-off point for creating a game is to change someone else’s.

“Make a mod,” advises Barnett. “You’ll start past the bigger technical humps, have an established audience and learn a lot from the existing code. Plan for something that you can do with one or two people in a month – that way it only bloats to a year or so of work. Get it out there as soon as possible to collect feedback and iterate on your ideas. It’s a great way to learn, get noticed and build a portfolio.” The increasing focus placed on usergenerated content by developers eager to engage with their communities – from Media Molecule’s commissioning of LittleBigPlanet 2 community level pack Hansel & Gretelbot and RedLynx’s inclusion of Trials HD players’ tracks in its Big Thrills DLC, all the way up to the free CryEngine 3 SDK – ensures that there are more of those opportunities to get noticed than ever before.

“Play the games you like and identify things that could be improved upon, things that could be expanded upon,” enthuses Kohli. “Then, try to make something that shows your process. Make a small Flash game or a mod – the easiest way to learn how to make games is to make them! “I think it’s great that there are so many new ways for people to make and play games. It can only be a good thing if more people play, and that in turn means that there are many more ways of getting into the industry.”