Overview
critter safARi
Conceptualizing a mobile app that augments zoo visits to drive impact on people and our planet.
This project began as a self-led, cross-functional student team project that I led and it evolved into a 1-week, AI-assisted solo design sprint.
Ultimately, this project explores how we can build a viable, user-friendly product that improves ecological empathy.
Role
Duration
2 months
Cycle 1 Team
Project
Tools
Figma (including Figma AI - Jambot), Kling AI, ChatGPT, Midjourney, Unity, Xcode, GitHub, Blender, Photoshop
CYCLE 1
CROSS-FUNCTIONAL TEAM PROJECT · 6 WEEKS
Problem Definition
Project Pitching
Early Prototyping (Unity)
Lightweight User Testing (A/B Test)
Catalyzing a Team Project
To a small class, I pitched a product opportunity that attracted 2 cross-functional peers.
Pitching the Opportunity
Visitors are often frustrated when zoo critters are not visible or are far away — so how can we balance animal care with visitor engagement?
This idea stemmed from a personal experience and was later affirmed by user research. I had once gone to a major zoo where almost half of the animals were missing, and I had also noticed that it was tough to understand the sizes of animals in large enclosures due to the sheer distance.
Opportunity Embodied
User research distilled as our MVP persona, Timmy.
12 y/o · Sloth Enthusiast · Older Brother
"Both of our [his and his little sister's] favourite animals were missing! Jamie and I had waited for our day at the zoo for so long."
Pitching the Solution
Mobile AR can bring animals close to the user — revealing unparalleled information about appearance and size, while fostering eco-consciousness via bit-size educational snippets.
North Star
This app was intended as an one-off experience for our local zoo.
Leveraging Unity to Inform Design
I built low-fi, Unity-based, interactive prototypes to decide between two interaction patterns.
Lightweight A/B Test Results
100% of testers preferred option B for the closer proximity between user action and feedback and the reduction in irrelevant copy.
On retrospect, to increase testing accuracy, I would match the copy in the pop-up window to the bottom sheet's "paragraph 3" as they intend to represent the same information.
Collaborative Achievement
Together, we created a functional prototype where users can scan markers to reveal critters.
In this cycle, I developed in Unity, established multimedia workflows, managed the project in Notion, and conducted lightweight UX research. Meanwhile, my teammates took ownership of UX/UI, branding, and 3D design. Our collective efforts earned a high course grade, excited the academic faculty, and was showcased to junior students.
CYCLE 2
SOLO DESIGN SPRINT (2 WEEKS)
User Interviews
Competitive Analysis
AI-Enhanced Workflow
LLM Feature
Design and Business Strategy
UX Copywriting
Deepening User Research
To evolve this app, I conducted user interviews and competitive analysis that pointed to opportunities in wayfinding, education, and marketing.
New Opportunities Embodied
Beyond Timmy as our primary persona, I distilled more personas and scenarios to guide design.
Madeline
Secondary Persona
38 y/o · Mother of 3 · Busy Professional
"As a family of five, it’s overwhelming — route planning, navigating, and meeting every child’s needs is no small feat."
Mateo
Tertiary Persona
66 y/o · Retired Business Owner
My knees wish we took the shortest route through the main exhibits. I also wanted more conservation stories — we’re always eager to learn how to build a better world for future generations.
Celine
Quaternary Persona
21 y/o · Student · Blogger
"Social media led me to visit the zoo today — my friend posted a super cute picture with the new baby fox!" "I wish there were cool events for people my age like night-time safaris…"
New "How Might We"s
Simplify route planning
Support deeper learning
Set up a viable business plan
Key UX Updates
To achieve the new HMWs, I added a LLM feature, changed the AR strategy, and enhanced business considerations.
OR
Opt. 2: LLM-Based Route Customization
(Low Constraints Context)
Cycle 2: SLAM (Simultaneous Localization
and Mapping)-Based AR
QR-Code Based Zoo Specificity
Early Testing
Enhancing my process with AI, I built this wireflow and tested with 5 potential users via think-aloud.
GenAI Art for Conceptualization
GenAI was a time-saver in conceptualization. Below shows how I edited raw AI outputs to fit the vision more precisely.
User Testing-Driven Iterations
User insights drove several improvements to increase engagement and efficiency.
User-Led Improvement
Switched to seasonal and changing cast of AR critters
Why This Works
Gamification that encourages zoo revisits
User-Led Improvement
Added selfie option
Why This Works
Improves social media presence and influence
User-Led Improvement
Reduced interactions in route selection screen
Why This Works
Speeds up task for users
Visual Identity
I found Midjourney especially helpful for exploring color palettes — and synthesized the rest manually.
Final Prototype
The final output is a concept of a user-friendly app that drives zoo engagement, offers bite-size education, and simplifies navigation — to ultimately foster ecological empathy in users.
LLM-Based Route Planning
Reduces friction for users like Madeline and Mateo with specific route needs and low tolerance for hassle.
Education via AR Critters
Applying the user-validated interaction B, we ensure critter visibility for Timmy.
Navigation Plus AR Access
The map enables access to GPS-based AR while further increasing navigation ease.
Next Steps
I'd take this concept further by connecting with various zoos to identify their unique needs and constraints.
Based on these insights, I would pitch a roadmap and gather a small development team — leaning on my Unity programming skills to bridge implementation and design. Further down the line, to gauge product success, I'd be interested in tracking metrics like retention and ARPU.
Reflections
Exploring GenAI as a Creative Partner
This project was an amazing opportunity for me to apply cutting-edge image/video generation tools like Kling AI, Midjourney, and ChatGPT as well as evaluate their competitors.
However, for launch-ready projects, I'm confident I'd be able to create visual assets manually instead to ensure full creative control.
In the future, I'd like to explore AI prototyping to enhance wireframing.
Valuing Transparency, Clarity, and Consideration
The team experience in this project taught me the value of clear and considerate communication and documentation to enable smooth cross-functional collaboration.
Sharpening Core Design Process
The rapid design sprint allowed me to reflect on my core design process and practice agile workflows.