Longhorn Loop
A personalized event board bringing 1,000+ UT orgs’ events into one feed.
- Categories
- Year
TL;DR
A personalized event board for UT students — everything happening on campus, sorted by what you actually care about.
- Role
- Design Fellow → Design Lead (Aug 2026)
- Timeline
- Aug 2025 — Present
- Team
- Product: lead + 2 fellowsDesign: lead + 2 fellowsEngineering: lead + 3 fellows
- Tools
- Figma, Linear, Notion
- Context
- Mobile · 0→1 · Consumer · Beta Sept 2026
- I designed
- Account creation and onboardingEvent detail and map overlayExplore filtersProfile and edit profileReporting flowAcademic standing updateCore components — buttons, dropdowns, search
Background
Campus events are everywhere except in one place
UT Austin has 55,000 students and over 1,000 recognized student organizations putting on events constantly. The events aren’t the problem — finding them is. They live on Instagram stories, GroupMe, and printed flyers, and while HornsLink exists as UT’s central org database, it isn’t where students actually look.
A student can miss the one event meant for them because it lived on an account they don’t follow.
Asking students how they actually find things to do
We ran a 30-response survey to find the patterns, then two rounds of interviews — about 20 students total — spaced across the project so we could check our assumptions twice.
Survey, 30 students: how students currently find campus events, and what makes them decide to go.
Interviews, round one, roughly 10 students: early in the project, to understand the discovery habits behind the survey answers.
Interviews, round two, roughly 10 students: later, against working designs, to test whether those assumptions survived contact.
Three things students needed before they’d show up
1. Tell me what’s actually in it for me. Students decide fast, and they decide on specifics — free food, swag, prizes, who’s speaking. A title and a time isn’t enough to justify the walk.
2. Show me why this event is for me. Relevance had to be visible on the card itself, not buried a tap away.
3. Make it findable, then remind me. Discovery was only half the problem. Students found events they wanted to attend and then forgot about them.
How might we bring campus events into one place and surface the ones that actually fit each student?
HornsLink is a directory — you go to it already knowing what you’re looking for. We were building a feed: something that knows what a student cares about before they open it, and that stays with them from finding an event to actually showing up.
Process
01 — Cutting a tab
The first structure had five tabs: Home, Explore, Create, Events, Profile. Events and Profile were doing overlapping work — both were places you went to see your own stuff.
We folded Events into Profile as three filters: Going, Saved, Posted. Four tabs, one mental model, and a profile that finally had a reason to exist beyond a bio.
02 — Onboarding that earns the personalization
The whole product depends on knowing what a student is interested in, so interest selection is the most important screen in the app. The first version was a 3×4 grid of unlabeled tiles asking students to choose three or more — no search, no structure, no way to find something specific.
The rebuild groups interests into six collapsible categories — Social & Networking, Gaming & Entertainment, Learning & Education, Sports & Fitness, Food & Drink, Nightlife & Parties — adds search across all of them, and shows a running count per category. Searching something we do not have returns “Tag not listed? Send it in”, so an unmet interest becomes a signal instead of a dead end.
Avatar selection got the same treatment. The lo-fi made students pick one of four preset Bevos to continue. The rebuilt version offers three paths — upload a photo, customize a Bevo, or skip — so a step that was a gate became one a student opts into.
03 — Designing around flyers we do not control
Orgs make their own flyers, and they make them at whatever aspect ratio they want — vertical, square, horizontal. The event board is mostly flyers, so the card had to hold all three without the feed falling apart.
I tested the variants against a two-column layout and locked the card to a fixed ratio with consistent metadata below the image: org name with verification badge, date, time, location.
04 — Designing for bad behavior
An open posting platform for 50,000 students needs a way to flag what should not be there. The report flow covers four reasons — violent or harmful, misinformation, troll or spam, concern not listed — requires a description, blocks submission until both are filled, and confirms that a human will review it.
Small surface, but it is the difference between a prototype and something you can hand to real students.
Final design
Campus, all in one place
A personalized board built from the interests a student picks during onboarding, with a time-aware greeting and upcoming events up top.
Find something outside your bubble
Explore is search plus filters: sort, time of day, in-person or online, distance, and Perks — free food, free swag, free stuff, prizes. That last filter came straight from the first finding. Students told us what makes an event worth attending, so we made it filterable.
Everything you need in one tap
Date, location, host, interest tags, who is going, and RSVP on one screen. Save it and the app reminds you before it starts — the gap where students told us they lost events they had already found. Tapping the location opens a map overlay with walking distance and a handoff to Maps. If the org has not posted a location yet, the event says so and offers to notify you when it is added.
Your events, three ways
Profile holds Going, Saved, and Posted, with search and filters inside each. Org accounts use the same layout with an Upcoming and Past split.
Built to survive a team that turns over
Fellows cycle out every year. The system is how the work stays consistent when the people do not. I built the button set, the year and classification dropdowns, and the search-and-dropdown pattern used across onboarding, explore, and profile.
The states nobody screenshots
The academic standing prompt is the one worth calling out: it asks students to update their year, so the feed stays relevant past freshman year.
Status
Where it stands
Longhorn Loop enters beta testing in September 2026. I joined at ideation in August 2025 as a design fellow and now lead the design team.
- Screens and states designed
- 70+
- Flows owned end to end
- 9
- Design fellows I now lead
- 2
- On the project
- 1 year, from ideation to beta
Reflection
Designing for a team that turns over
Fellows rotate out yearly, so anything undocumented gets rebuilt or quietly dropped. Building components and states other people could pick up mattered more than any single screen I made.
We were also our own users, which cut both ways. It made the research fast to interpret and it made it easy to assume we already knew the answer — the second interview round existed partly to keep us honest about that.
From making screens to owning them
As a fellow I was responsible for my own flows. As lead I am responsible for whether all of them add up to one product — which turns out to be a mostly different job, and mostly about running critique well.
The open question
A 30-person survey and two rounds of interviews got us to something coherent. What they cannot tell us is whether the personalized board actually works — whether a student opens it, finds an event they would not have found otherwise, and goes. The September beta is the first real test of that, and it is the number I want.