Skip to main content
Back to top

DigiPen graduate Tejeev Kohli (TK) talks with us about working at Valve, about the courses that helped prepare him for it, and about the importance of playtesting.

Which degree program did you graduate from?

I graduated from the RTIS program at DigiPen in April 2009

Where are you working currently?

Valve in Bellevue.

What is your role there?

I am working as a level designer and programmer on Portal 2.

How did you go about applying for/securing this job?

I was displaying my game Tag: The Power of Paint with the rest of my team at the DigiPen career day in spring 2009. There were some people from Valve there that we had a long discussion with about our game, and the way we went about creating and refining it. We were showcasing our game at GDC that year as part of the IGF competition. At GDC, a few more people from Valve came by and looked at the game, and liked it. After these encounters, we had a meeting with them and they offered jobs to our entire team.

Do you think attending and graduating from DigiPen helped you secure this job? How?

Definitely. At DigiPen we learn to work as part of a team from freshman year onward; that is an invaluable skill. We also learn to make complete games from scratch every year, so by the time I was done with school I had already worked on four different games with different teams. Even though the games that I worked on were small student projects, working on several from start to finish as part of different teams is something you don’t get at other colleges.

What DigiPen course do you find the most applicable to your current job?

The classes that helped the most were probably Game Software Design and Production projects (GAM) classes. Going through the process of designing and then building a game from scratch helped us understand the amount of work that goes into creating even a small student project. The things I learned in my GAM classes helped me appreciate the amount of work that goes into creating a large game like Portal 2.

A lot of the programming work I do here is gameplay oriented, so I also use a lot of stuff I learned in my Math and Computer Science (CS) classes. Linear algebra is used almost every time I have to program something and it doesn’t hurt to know some of the graphics-oriented algorithms from CS200 and CS250.

It’s our understanding you have a big title coming out, Portal 2; what can you tell us about it?

It’s really exciting to be working on such a big game with all these great people at Valve. We had a great showing at E3 last month and it’s great to see that a lot of people are looking forward to the game. Unfortunately, I can’t tell you much more about the game. We have released some videos that showcase some of the different mechanics in the game, you can see it at:

DP: Did working on your game projects help prepare you for working on this game? If so, how?

We did a lot of design work on Tag before we actually started making the game but once we started building the game we quickly realized that the design we had come up with had too many holes in it and was probably not going to be that fun to play. We went back to the drawing board and tried to re-work our existing design, but once we had a working prototype of that design we realized again that it was probably not going to be very fun.

We went through this re-design process a few times until we understood that the way to further our design was by playtesting the game. The idea of letting playtesting drive the design of the game forward is something that we learned very late in our development process, but once we learned how valuable this tool was, we playtested our game whenever an opportunity presented itself. We would all watch the person playtesting the game and take notes while they were testing. We would then use the playtester’s feedback along with the notes that we took to iterate on all the mechanics and puzzles in the game.

Valve works very similarly as far as playtesting is concerned. Every new mechanic or puzzle that goes into the game is playtested by several people before we know for sure if it’s going to stay in the game or not. Everything in the game is iterated on, based on the playtesting feedback that we get. Working on Tag helped prepare me and the rest of the team for this sort of a design process.

Did you adapt any game concepts that you created in your 4 years here for Portal 2? If so, what and why did you choose to adapt it for this particular game?

We have adapted certain aspects of the paint mechanic from Tag into Portal 2. When we first started at Valve we discussed the idea of using paint along with portals and began working on a prototype of that. After some experimentation and playtesting we found that the two mechanics worked really well together and seemed like a natural fit.

If you could give one piece of advice to DigiPen students what would it be?

Playtest your game! As I said, we learned it really late in our development process while working on Tag, but even doing it for the last few months helped to completely transform the look and feel of the game. The most important piece of advice I could give would be to playtest early and playtest often. As soon as you have a vaguely playable prototype you should grab other students and make them play your game. At first it will be really frustrating because they will play the game in the worst possible way and do nothing in the way that was intended. As you move forward and keep playtesting, things will smoothen out and you will be able to identify the parts of the game that aren’t as polished as you think they are. Identifying these unpolished areas early on will make your game better in the long run.