I designed and tested a photo-op discovery feature for GramCity, a fictional photo-editing app. Starting with a provided research brief, I built a map-first prototype to help people find nearby spots and evaluate them through reviews and visitor photos.
Solo Design Sprint
Brief to tested prototype
Core Flow Completed
Also said they'd use the feature
Testing Changed the Design
Full-screen photos to evaluate spots
The assignment and constraints.
A self-directed five-day sprint testing one thing: how quickly I could turn a provided brief into a credible, user-tested solution.
GramCity is a photo-editing app. The brief: add a way for users to discover where to take great photos.
The five-day sprint.
Five days to move from a provided brief to a tested prototype. Each day focused on one decision that moved the concept forward.
Cluster research, define user behavior, frame challenges.
Prioritize features, study rivals, ideate fast, by hand and with AI.
Storyboard the flow, build the HiFi screens.
Wire the screens into a clickable flow with Claude Design.
Test with five users, find friction, recommend fixes.
What people wanted to photograph
and how they chose locations.
Using the provided interviews, I clustered recurring behaviors and needs into themes I could design against:
The synthesis translated into three requirements for the experience:
Selecting a map-first approach.
I narrowed the concept around one core task: help people find a nearby photo-op, understand whether it’s worth the trip, and get there.
The insights pointed to one core experience: location-based photo-op discovery. Map/list switching and reviews + photos supported that decision, while search, saving, and filters stayed secondary.
I studied familiar discovery patterns so the sprint could spend its limited time solving the photo-op problem instead of teaching users new interaction models.
Browse without losing your place.
Jump straight to the type you want.
Proof that a spot's worth the trip.
I used Claude Design to explore multiple structural directions quickly, including map-first, list-first, category-led, and conversational concepts. I chose the map-first direction because it best supported nearby discovery while keeping the list one tap away, then translated that decision into a hand-sketched storyboard.
Exploration: AI-assisted concept generation
Committed storyboard: hand sketched, map-first
From flow to clickable prototype.
I mapped the core discovery path before designing screens so every interaction had a clear next step. The core flow moved from location access to discovery, evaluation, and directions.
The critical path: enable, browse, decide, get directions.
Explains why location improves discovery before requesting permission.
Photo-ops appear as pins, with filters and list view available without losing context.
Ratings, reviews, directions, and visitor photos help someone decide whether a spot is worth going to.
Once the core screens were defined, I used Claude Design to turn the flow into an interactive prototype. I directed the taps, toggles, and transitions around the critical task so testing could stay focused on one question: can someone find a photo-op, decide if it’s worth the trip, and move toward it?
Testing the core flow with five users.
I asked five participants to find a nearby photo-op, evaluate the location, and move toward visiting it.
All five participants completed the core discovery flow and said they would use the feature. Their feedback also informed a refinement to photo browsing.
Participants wanted to inspect other visitors' photos more closely before deciding whether a location was worth the trip. The existing grid gave them the proof they wanted, but not at a useful enough scale.
Fix: Make the photo grid tappable and open images in a full-screen, swipeable viewer.
The final photo-op discovery flow.
The prototype connects nearby discovery, map and list browsing, and location details to help people choose where to take photos.
How the entry screen communicates the benefit and gives people control over their settings
Clear payoff before permission.
Two views, one discovery flow.
How the two views show the same nearby locations without restarting the search.
How ratings, distance, reviews, and photos help people evaluate a spot.
Proof before commitment.
What this project
taught me.
Five days forced every feature to justify itself. Prioritizing the core discovery task kept the solution focused.
Map and list switching, category browsing, and reviews used familiar interaction patterns so people could focus on finding a photo spot.
Testing revealed that participants needed larger visitor photos to evaluate a location before visiting.
Claude Design accelerated exploration and prototyping. I selected the direction, interaction model, and final tradeoffs.
I'd run my own discovery round to pressure-test the provided research, build out the saved and plan-ahead flows, and re-test the photo-viewer refinement with users.
