The Engine Behind Every Supercell Game (and Why You’ve Never Heard of It)

If you know Supercell, you know us through our games. Clash of Clans. Hay Day. Clash Royale. Boom Beach. Brawl Stars. You might know we're organized around small, independent game teams. If you're really deep in the rabbit hole, you might even know about the champagne-for-failures thing. But you almost certainly don't know that every one of those games runs on a single homegrown game engine, built and maintained by one of our teams inside Supercell called Game Tech.

Game Tech is our central tech team: the people behind Titan, our proprietary game engine, along with a platform team that handles services used across all our games, and a layer covering cloud, security, and AI. The engine side is the focus here, but the team is bigger than that.

We've never really talked about any of this. That's partly because we've always wanted to be known for our games, not what they’re running on. But it also means a lot of really good engineers have no idea this work exists. We'd like them to know, because we're hiring and the work is genuinely unusual.

Edmond Maillard, who leads the editor and tools team on our Titan engine, almost didn't apply. He was looking at a few other European studios. He had to be nudged by a headhunter. He went looking for content about Supercell's tech side and couldn't find much. Then he had one interview with the lead of the Titan engine team. "He sold me the vision and I was like, I absolutely need to work here. I need to be here yesterday." Edmond is, in a way, the awareness problem with a face: someone who nearly slipped through because the content didn't exist. He's been here two and a half years now and still can't quite believe more people don't know.

We'd rather not miss another Edmond.

What does Titan actually do?

Edmond reaches for Formula 1 to explain it. “The game teams are the drivers and pit crew, performing under pressure on race day. The game is the car. Game Tech provides the components. Reliable, tested pieces the teams can build with. Game teams want to focus on being creative. They just want to think, 'I want to make cool stuff.' So we offload all those problems from them.

Titan is built specifically for mobile, and it targets the full range of devices, not just the flagship phones. Think: the cheapest, oldest Android handset someone somewhere in the world is still playing Clash Royale on. When Supercell says "as many people as possible," that translates directly into an engineering constraint: Titan has to run beautifully on hardware many studios write off.

Building for mobile from the ground up means the team avoids a huge amount of the optimization work that other studios face when they use Unity or Unreal on mobile. As Edmond puts it, "By specializing, you allow yourself to avoid a lot of problems entirely." You trade those problems for a different set, and you get to be very, very good at that set.

How good? When mo.co shipped last year, it launched on a new rendering backend that made it only the second game in the world at that level of rendering support, the other being a miHoYo title. Someone on the team recently developed a novel algorithm for 3D model compression that outperforms what's available in open source. This isn’t maintenance work. This is the frontier.

What went wrong (and why that matters now)

Here's where it gets honest.

Timo Tervola heads the Titan engine team. His background is unusual. Management consulting at Bain, corporate strategy at an industrial company, then recruited to Supercell to help manage the sale of its majority stake to a new buyer. He ended up leading Game Tech's transformation after uncovering, through hard data, a problem nobody had quantified.

Supercell's early philosophy was: just build the next feature. Don't over-engineer. And it worked. Other game studios went bankrupt building too much tech too early. But when you keep building on top of early, hacky solutions year after year, you accumulate debt that's almost invisible. Timo calls it "a silent killer." Games still work. Players still play. But performance degrades subtly on average devices, signals so weak that no player could articulate them and no feedback form could capture them. The truth was in the data.

When Squad Busters launched, the accumulated debt became impossible to ignore. Timo ran the numbers and found that the problem wasn't Squad-specific. It was company-wide: over a decade of accumulated tech debt that was quietly degrading the experience on ordinary devices. "It hit us in the face," he says. That lit a fire inside the company.

Game teams had already been telling them directly. Timo recalls the feedback: we like you guys, but we can't trust your roadmaps, because every time we make plans together, fires pull you away. That kind of honesty stings. It was also exactly what was needed.

Timo's assessment is blunt: "We had not invested enough." No spin. The good news is that the reckoning created the mandate to actually fix things, and the ambition to build something much bigger.

From firefighting to building

To understand the shift, you need to know what came before it.

For years, we kept headcount deliberately tiny. Timo tells a story about Frank, the GM of Brawl Stars, who said on stage that it was easier to approve a million-dollar invoice than to ask for one extra person. That was the culture: small teams, big output.

Around 2020, that changed. Our CEO, Ilkka, went on stage and said that our best days were ahead of us, that players wanted more content, faster updates, and we needed to deliver. Game teams got the green light to grow from 20 people to 50 or 70. Great for game teams, but it also meant the tooling underneath, built for teams of 20, suddenly needed to work for teams nearing 100 people. That gap became the opportunity. Game Tech was stretched thin, stuck putting out fires, and running on one-person projects that were reactive and small in scope.

That's shifted. Timo got every game team in a room, started with about a hundred priority items on a wishlist, and narrowed it to eight focused areas over a three-year horizon. The mindset went from "let's just build the next thing one game team asked for" to genuine long-term planning, something that had never happened at this scale before.

Edmond sees it from inside the engine team. When he joined, the Titan team was about 20 people. It's approaching double that now, which sounds like a lot until you remember this team's work reaches hundreds of millions of players. That's extraordinary leverage, and the ambition is finally matching it. The team has started running its first real priority-project reviews. They're delivering on promises. It's not a paper exercise anymore.

How decisions actually get made

One thing that surprises people: Game Tech doesn't get to force its tech on game teams. Teams choose whether to use it.

This surprised Edmond too. He came in with preconceptions from commercial engines and expected to push familiar approaches. "I learned very quickly that was not the way people wanted it to work. So I had to change my mindset, but change it in a way that in the end felt better." The game teams know their games best. Game Tech's job is to build something so good that adoption is a choice teams want to make, not one that's imposed.

When Edmond noticed artists across the company struggling with inadequate tooling, he flagged it. His lead asked what he'd do about it. Edmond said they needed a team. The answer: do it. "People trusted me: you identify the problem, you have an idea of how to solve it, go ahead, try it." He built that team from scratch. What could have been a six-month approval process was a single conversation.

Timo's principle for how the growing team stays coherent: "closely aligned, loosely coupled." Teams share a mission but own their domains completely. Cross-team projects need real buy-in, not top-down mandates. If a team doesn't feel something is important, nothing happens, and that's by design. Forced delivery without buy-in creates worse problems later. For engineers who've spent careers being told what to build by people who don't understand the work, that might sound almost too good. Timo's version of it: "I thought before that I used to work in places where I was given a lot of responsibility, but I was wrong. I just didn't realize what it could look like."

Is this for you?

The short version, from Timo: Game Tech is for someone who loves to execute, loves to own things, wants their work to reach hundreds of millions of players, and genuinely loves games.

The longer version involves what you'd actually be doing for people. Game Tech engineers build the tools that artists and designers depend on. When those tools are great, creative people can push boundaries. When they're not, they suffer in silence, and Edmond saw this firsthand. His personal mission now is to make Supercell the best place for artists to do creative work, and he's honest about where that stands: "Artists are depending on us to do a better job so they can express their creativity. We're far from that at the moment, and that's a failure. We have to solve that."

If that sounds like a problem worth solving, you might be who we're looking for.

That honesty is core to how we work. Ownership of failure is where trust gets built, something we take seriously enough to have written about it at length. Edmond mentions that they've turned away candidates who looked great on paper but didn't embody that. And on the other side: Edmond says that at previous companies he was often the most experienced person in the room, which sounds good but limits how much you grow. Here, he's learning from everyone. The caliber of people changes what's possible.

Come build with us

If what you're looking for is impact, consider the math. Changes made in Game Tech can touch every game and every player. As Edmond puts it, "In terms of impact, it's absolutely insane."

We're hiring across the engine team – explore our open roles here!

More from the people inside Game Tech coming soon.