AniVerse — An Anime Discovery Interface
AniVerse is a working iOS prototype for browsing, exploring and saving anime. I designed the interface in Figma and built it using Swift, UIKit, Storyboards and Core Data.
Duration
February - April 2026
Tools
XCode, Swift, Figma
#ios_devlopment
#prototyping
CONTEXT
Building beyond the brief
The project began with four required views, an XML data source and a Swift model. That gave me the structure, but not the experience between each screen.
I chose anime because it gave me rich artwork and detailed content to design around. My main task was deciding what belonged on each screen, how users would move between them and how their actions would remain consistent throughout the app.
I was also learning Swift and UIKit while building it, so design and development happened side by side. There was no final Figma file followed by a neat handoff. I would design something, build it, see where it broke down and return to Figma with a better understanding of the problem.
Figma rarely objected to an idea. UIKit had stronger opinions though!
WORKING IT OUT
Making the catalogue useful at a glance
The catalogue needed enough information to help someone judge a title without overcrowding every row.
I used Figma to explore the visual direction and rough out layouts, but a lot of the iteration happened once those ideas were inside Xcode.
I kept the poster, rating, season count, episode count and runtime visible while moving genres, audio and longer descriptions into the detail views. The decision was less about fitting everything in and more about deciding when each piece of information became useful.
In the Xcode playground, I had to deal with actual table views, view controllers, navigation and the way those pieces behaved together with code.
So the process became fairly back and forth, iterating between designing in Figma, building in the Xcode playground, testing with the iOS simulator and changing around code.
Sometimes building the design revealed practical issues that were not obvious in Figma, so I adjusted the layout or interaction. In other cases, I learned more about SwiftUI so I could implement the original design more accurately. That was probably the biggest difference from the way I normally work, and one of the steepest learning curves of the project.
THE INTERFACE
Giving each screen a job
Titles could be saved from the catalogue or detail screen, so the same state had to remain accurate across both views and the Saved tab.
I used Core Data to persist favourites, then designed the surrounding states: an empty collection, filled and unfilled hearts, swipe-to-delete and a confirmation before removal.
What looked like one small interaction in Figma became a lesson in keeping several screens and the underlying data in agreement.
The Add and Edit screens combined text, genre selection, numerical details, links, ratings and poster images.
I gave both screens the same structure, grouped related fields and used familiar patterns such as selectable genre chips and star ratings. Building these controls in UIKit forced me to think beyond their appearance: each one needed a selected state, an existing value and a clear response when changed.
OUTCOME
The project that taught me to design closer to the code
I completed a functional prototype covering catalogue browsing, add and edit flows, persistent favourites, deletion, audio playback and in-app web content.
Building the app made implementation part of the design process. It showed me where a polished screen could still be hiding unanswered questions about navigation, behaviour and data.
AniVerse wasn't a research-heavy project, but it taught me something that has been useful in everything I've worked on since: the technology you're designing for can change the design itself.
I want to keep deepening my understanding of code and how design decisions carry through into development, particularly as AI-assisted workflows and "vibe coding" make it easier for designers to build and test their own ideas.