Fizz

Fizz is an anonymous social platform for college students—real-time feeds tied to specific universities where students can share their ideas freely and find community without judgement. I joined as a Product Designer in summer 2025.

Context & challenge

Anonymous platforms lower friction to post but reduce user's investment that drives retention. When I joined, Fizz had product-market fit on dozens of campuses with thousands of users across the country but lacked a consistent design system, coherent navigation, and any monetization surface. The mandate: bring craft and system thinking to holistically mature the product without harming the trust of user's wanting a home to share their voices unfiltered and unjudged.

My role

As the sole designer across the full product, I set aesthetic and functional direction, prototyped interactions, and measured design impact of shipped features and improvements. I worked closely with engineering and product management to collaborate on solutions and weigh priorities as the platform grew.

I also contributed code to our codebase to polish interfaces and fine-tune interactions using AI-tools like Claude Code.

Core Navigation Improvements to Profile and Activity

Problem

Tabs were inconsistently placed, the information architecture didn't reflect how users moved through the app—especially in relation to direct messaging being hard to reach and off screen when scrolling the feed, which is one of the apps highest-engagement surfaces and a strong contributor to retention.

Approach

I audited navigation patterns and session recordings, then prototyped multiple tab architectures. Key decisions:

• –Elevate DMs to a persistent primary tab
• –Reorganized the tab bar around actual usage frequency
• –Introduced a unified "activity" hub for notifications while clearing up the mental model of the user's profile

Navigation schema before and after I joined Fizz

Outcome

DMs increased 37% from reducing friction to the feature, not from changes to the messaging product itself.

Good design decisions compound. These results reflect work that was scoped, prioritized, and iterated on with a clear hypothesis—not shipped and hoped for.

Rapid prototyping as the primary communication tool. Working prototypes—not decks—were how I aligned engineers and stakeholders. A testable concept surfaced implementation constraints faster than a spec document.

Data as a design partner. I treat measurement as part of the design process, not a post-launch check. Every significant decision starts with a hypothesis about what will move and why—so the outcome, whatever it is, teaches something.

System thinking at component level. Working alone across a full product required every UI decision to account for its implications elsewhere. Speed and quality are not opposites if you know which decisions are foundational and which are reversible.