The problem hiding in plain sight
A single button. Seven different versions. Seven different files. No reliable way to know which one was current, which one was correct, or which one a developer had actually implemented last week.
That was the reality I walked into as Lead UX/UI Designer at a fast-growing tech company. The symptom was visual inconsistency. The cause was structural: there was no shared design language, no single source of truth, and no common ground between designers and developers trying to build the same product. Teams were miscommunicating at every handoff. Developers were rebuilding what designers had already solved. Designers were answering the same questions they'd answered three sprints ago.
The company needed more than a cleanup: it needed a system.
NDA applies: visual resources partially replaced.
What I did
I led the design system end-to-end: research, information architecture, component design, documentation, and cross-team rollout; grounded in two structural decisions made before a single component was designed.
First, Atomic Design methodology as the organizational framework: atoms → molecules → organisms → templates → pages. A hierarchy that made the system navigable for designers and legible for developers, without requiring anyone to hold the whole thing in their head at once.
Second, the Fibonacci sequence as the mathematical foundation for spacing and proportional harmony, not an aesthetic choice but a systematic one. When spacing follows a consistent mathematical logic, visual decisions become defensible rather than subjective, and the system scales without breaking.
But before any of that, I ran research with the people who would use the system every day. User interviews with designers and project managers to map their actual workflows and frustrations. Contextual inquiry: watching designers navigate existing assets in real work conditions, not described conditions. Surveys to quantify satisfaction gaps and prioritize what mattered most.
The system was designed for two primary users: designers who needed to find and implement components without interrupting their flow, and project managers who needed reliable documentation they could trust when a deadline was on the line.
The unexpected obstacle
The hardest part wasn't designing the components. It was getting the team to trust that this was worth the disruption.
Asking designers to stop using their own files and adopt a shared system mid-project felt like asking them to change shoes while running. Adoption required more than good documentation: it required the system to be so clearly better than the alternative that resistance became friction people chose not to maintain.
That's what shaped the rollout strategy: starting with the components teams touched most often, proving value quickly in contexts they already cared about, and letting the documentation speak for itself before asking anyone to change their habits wholesale.
Ideation, Sketching & Wireframing
Brainstorming: Use of the Fibonacci Sequence and Atomic Design
Listing and Sorting Components
Components grouped by Atoms, Molecules and Organisms
Paper Wireframe
Digital Wireframes
Key Mockups
The Outcome
The signal came unsolicited, from a developer on the team: "We get something from the library and it's done in a snap… faster than expected is phenomenal." That's the measure of a design system that works: not how beautiful it is, but how invisible it becomes.
The system eliminated 7× component duplication and became the single source of truth across 10+ applications. Developers stopped waiting on designers for answers they could find in the documentation. Designers stopped rebuilding components they'd already solved. Cross-team collaboration accelerated because everyone was finally speaking the same visual language... and the visual language was written down.
What I carried forward
A design system is only as valuable as its adoption rate. You can build something architecturally perfect and have it ignored because the rollout didn't account for how people actually work. This project taught me to treat adoption as a design problem (with its own user research, its own friction points, and its own success metrics) not an afterthought to the system itself.
Tools: Figma, Atomic Design methodology.
Role: Lead UX/UI Designer: concept to launch. Owned research, IA, design, documentation, and cross-team rollout.
Built in 2024 for internal use. Not publicly available due to client NDA.
