Que
A simpler way to save.
"What if you never had to remember where you saved something?"
Saved in the moment. Found in seconds.
No new apps to open. The standard Share button is the entry point. The second the inspiration hits, it's captured — without interrupting what you're doing.
Que reads, categorizes, and timestamps every save in the background. By the time you look for it, it's already organized.
Type what you remember, not what you saved. 'That red jacket from last winter' is a valid search. The right result comes back.
You saved it. You lost it.
"I saved it. I know I did. But sitting in my car, I couldn't remember the title, the app, or even where to start looking."
The moment you hit "save," your brain lets go. That's the problem.
Saving isn't finding.
7 in-depth interviews, 1 shared friction: Users remember the act of saving, but lose the path to retrieval.
Titles are forgotten; context lingers.
We identified 7 Memory Anchors
that outlast any file name. These became the structural foundation for our search.
1. Save everything. Find nothing.
Bookmarks without organization are just dead ends.
"I have 25 pinned tabs right now because if I put them in bookmarks, I'll forget they exist. My saves are a total mess."
2. Memory isn't text.
Users recall colors and time. Search bars demand exact titles.
"If I forget the title, I can usually only remember the main color of the image, or 'what was that thing I looked at yesterday evening?'"
3. The iMessage Workaround.
Users text links to friends—not to share, but to build a searchable timeline.
"I forward the ones I actually want to make to my roommate on iMessage so we can find them in our chat history."
Designed in Figma. Built in Swift.
Bridging Design & Code
Static mockups don't test retrieval. I built a fully functional alpha in SwiftUI to prove the search logic worked in practice.
Invisible Context Scraping
When a user hits Share, an engine runs silently in the background — extracting metadata, visual anchors, and body text. The complexity is invisible. The result is a memory you can actually find.
Capture, connect, retrieve.
One flow from start to finish. Que intercepts content at the moment of saving, organizes it silently, and makes it retrievable in plain language.
First test. Hard feedback.
The first prototype required users to manually sort every save into folders. That single friction point broke everything.
Critical Feedback:
- Flat Information Hierarchy: Users felt overwhelmed when forced to manually assign categories during the "save" action.
- High Interaction Friction: Leaving Instagram/Safari to save a link broke their natural browsing state.
The fix was removing choice. Zero-friction capture through the native Share Sheet, retrieval through natural language.
Real users. 12 hours later.
Users saved 10 items in the morning. Twelve hours later, they retrieved them using only vague descriptions. The AI architecture delivered. Then I pushed further — two extreme scenarios to test where the semantic logic breaks.
"How can I see through thick smoke?"
Twenty pages of infrared drone physics. A layman's question. Que bridged the gap — connecting the vague query to the precise scientific concept ("LWIR sensors") without any keywords in common.
"Find the laptop for 8K editing on a beach."
MacBook Neo specs as the source. "Editing on a beach" as the query. Que identified that the environment requires peak brightness and thermal efficiency — without those words appearing anywhere in the document.
Meet Que.
A Silent Copilot.
"Building Que taught me that technology should not replace memory, but act as a silent extension of it. The real complexity lies in designing it to be effortless."