Fuse Games kept its core team small on Galactic Racer, using external partners for everything from defined features to close co-development.
A small core, with capacity around it
Star Wars: Galactic Racer was built by a lean core team backed by external developers. Fuse Games CEO Matt Webster says that model helped the studio avoid the churn of repeatedly scaling up and down for individual projects. His account offers a practical view of how a smaller team can work on an ambitious game: outside partners can take on tasks ranging from clearly bounded deliveries to work embedded in the studio’s daily process.
In an interview with GamesIndustry.biz, Webster said Fuse’s average headcount was “a lot smaller” than at a large studio such as Rockstar Games. The comparison is useful as a description of scale, though it doesn’t tell us how many people worked on the game in total or how much work each partner handled. What it does make clear is that Fuse chose to keep its central group compact while relying heavily on outside contributors.
Webster said the studio wants to “operate at Premier League level” without repeatedly expanding and contracting its workforce. Downsizing, he argued, affects people and removes accumulated expertise that can be costly to rebuild. That’s a familiar production problem: project needs change, but teams also carry knowledge about tools, decisions, and working relationships that doesn’t return instantly when a new group is assembled. Fuse’s answer is to maintain a stable core and bring in partners as the work requires them.
Three levels of external development
Fuse divides distributed development into three tiers, based on how closely outside contributors work with the in-house team. At the most distant level, a partner receives a brief, delivers an asset or other output, and Fuse integrates it into the game. That arrangement can make ownership clear, but a written scope alone won’t guarantee that the result fits the project’s needs. Webster stressed that even clearly scoped work depends on a strong relationship: “You get a better result when you do so.”
The middle tier covers self-contained features assigned to an external team. Webster cited Galactic Racer’s photo mode as an example. Because the feature could be defined in advance and built as a distinct unit in Unreal Engine, a cross-disciplinary group could develop it and Fuse could slot it into the wider game. The appeal is practical: a feature with a defined boundary can be handed to a team with the needed mix of skills, while the studio retains responsibility for how it fits into the finished product.
“You need production, art, UI, and code to create that thing,” Webster said. The point is that even a feature with a clear scope doesn’t belong to one discipline alone. Production has to coordinate the work, while art, interface design, and programming all contribute to the result. That makes the handoff more involved than simply ordering a finished asset, but it can still be managed as a distinct piece of development.
The closest arrangement is co-development, where external contributors take part in the studio’s day-to-day work. They join stand-ups and Slack channels and have access to the code base and team processes. Webster described them as “an extension of our team,” despite being employed by another company. Here’s the thing: the closer a partner gets to the core workflow, the more the arrangement depends on shared context and communication, as well as technical access.
A flexible production model for a released game
Star Wars: Galactic Racer launched earlier this month on PC, PlayStation 5, and Xbox Series X/S. In our review, the game scored 9 out of 10, with praise for its racing mechanics, takedowns, and roguelite campaign structure, which supports varied vehicle builds. A review score can describe the reception of the finished game, but it doesn’t reveal how production was divided. Webster’s comments provide a separate view: the studio treated staffing structure as part of the production approach.
Webster has also discussed players’ ability to handle cognitive load and why racing games need to offer a challenge. For Fuse, the production takeaway is just as specific: keep expertise in a steady core, then match each external team’s involvement to the work it can own. That’s a clear description of the model, though its success still depends on defining responsibilities and coordinating the people doing the work.
What the model means for other studios
This is more significant than it looks. A studio’s headcount is only one measure of its capacity; the way it assigns and integrates work also shapes what that team can deliver. Fuse’s three tiers offer a useful distinction between buying a defined output, commissioning a contained feature, and bringing outside developers into daily production. Those arrangements ask for different levels of coordination, and treating them as interchangeable would obscure the practical choices involved.
If you’re working in this space, the model raises a direct question: which expertise needs to stay inside the studio over time, and which work can be assigned to partners with a clear boundary? The answer will vary by project. And yet, the logic is straightforward: a compact core can preserve knowledge, while external teams add capacity where their skills match a defined need. The difficult part is making those boundaries hold when a feature connects to the rest of the game.
(And this is the part most people overlook.) Calling external developers an extension of the team describes a working relationship, not just a staffing arrangement. It implies access to communication channels, code, and established processes, all of which can help contributors work within the project’s existing structure. It also makes coordination part of the production cost. A partner who receives a brief has a different role from one joining stand-ups and working directly in the code base.
Webster’s account doesn’t establish that this approach will suit every studio or every project. It does make a case for matching the depth of collaboration to the work, while avoiding repeated changes to the core team. The test is whether the studio can preserve useful expertise and bring in the right support without making integration a bottleneck. A lean team, with partners assigned the right scope.
Discussion
Sign in to join the discussion.