Fujitsu Launchpad (FLP)

Fujitsu Launchpad (FLP), Fujitsu’s venture studio, builds and scales new digital ventures through lean product teams working across multiple web and mobile projects. With a small team shifting between initiatives and tight delivery timelines, FLP needed a fast and practical way to keep Flutter codebases aligned, catch issues early, and reinforce shared engineering patterns without adding process overhead.

DCM became a key part of that workflow, helping the team maintain code quality locally and in CI while supporting a venture model where speed and consistency both matter.

"We just wanted something that would solve our problems, work fast, and help us keep quality high."

— Andrey, Mobile Chapter Lead, FLP Studio

Challenge

FLP operates in a delivery environment where mobile teams need to move quickly, validate ideas early, and reuse proven approaches across projects. That creates a different kind of engineering pressure than in larger organizations.

Quality still matters, but teams do not always have the luxury of dedicated platform support or long cycles to refine internal tooling and standards. Before DCM, the team relied mainly on Flutter’s default linting and analysis tooling.

That covered the basics, but it did not fully address some of the problems they cared about most.

One clear gap was unused code detection. The team found it surprising that there was no straightforward way to identify unused classes or functions directly in their regular workflow. At the same time, FLP wanted more than just basic linting.

Because they build multiple applications and share patterns across projects, they also needed a way to guide developers toward approved internal abstractions and discourage framework usage that did not align with their desired application build practices.

Tooling speed was also critical. Any solution had to run efficiently on developer machines and in GitHub Actions to support day-to-day development without becoming a bottleneck.

"It was quite surprising that there was nothing built in Dart that could simply tell you: you don’t use this class or this function."

— Andrey, Mobile Chapter Lead, FLP Studio

Solution

FLP adopted DCM as the primary static-analysis layer across its Flutter projects, integrating it both on developer machines and in CI via GitHub Actions. The team uses DCM for linting, unused code checks, unused file checks, and validating assets against project-specific requirements.

That gave them a broader and more practical quality layer than the default analyzer alone, while still fitting naturally into their existing development process. One of the most valuable capabilities of DCM has been the ability to enforce internal engineering patterns through rules such as banned usages.

This allows the team to guide developers away from direct access to certain APIs and toward the abstractions FLP wants teams to use instead. For example, rather than letting developers access theme values or framework behaviors in inconsistent ways, FLP can steer them toward internal style tokens, shared abstractions, and preferred access patterns. The same applies to framework interactions such as dialogs and other services, where the team has created higher-level abstractions and wants developers to stay consistent with them.

Instead of relying only on code review comments to correct those patterns later, DCM helps encode those decisions directly into the workflow. DCM also helps FLP keep standards synchronized across multiple projects.

As new versions are introduced, teams can gradually adopt updated checks, use baselines where needed, and keep moving without creating unnecessary disruption.

"It helps us keep developers aligned with the way we want applications to be built."

— Andrey, Mobile Chapter Lead, FLP Studio

Impact

With DCM in place, FLP has maintained greater consistency across multiple Flutter projects while preserving the speed required in a venture studio environment. The most immediate impact has been faster feedback cycles during development.

Developers are now able to catch and resolve issues earlier in the lifecycle, reducing the need for repeated review cycles and late-stage fixes. DCM runs quickly both locally and in CI, making it easier for developers to catch issues early and fix them before they spread across projects, reducing rework and minimizing delays in delivery.

DCM has also become a practical way for FLP to scale engineering standards consistently across multiple projects. Instead of re-explaining preferred patterns from one codebase to another, the team can use static analysis to reinforce them consistently. That is especially useful in a setup where developers may shift between ventures and need to work within the same conventions.

Just as importantly, DCM gives FLP a way to maintain code quality without introducing a heavyweight process. In a model built around fast validation and delivery, that balance is critical: move quickly, but keep foundations strong enough to reuse and scale. In practice, this translates into fewer recurring code issues, more consistent implementations, and faster onboarding for developers joining new projects.

While FLP prioritizes speed, DCM helps ensure that this speed does not come at the cost of maintainability, enabling teams to deliver quickly while keeping codebases clean, consistent, and scalable over time.

"It works fast and efficiently. That core functionality, being able to check things quickly, works really well for us."

— Andrey, Mobile Chapter Lead, FLP Studio