Looks like you’re on a small screen. This case study needs a bit more space to shine ❤️🔥
Try again on desktop. It’s worth it, promise.

(And seriously, Who reviews project details on mobile anyway?)
Web App

Y2K Finance

Decentralized finance (DeFi) platform offering hedging mechanisms for digital assets.

Project definition

Y2K is a decentralized finance (DeFi) platform offering hedging mechanisms for digital assets. While its early-stage pixel-art UI helped capture community interest, the interface struggled to scale alongside the product.
As the platform expanded with advanced financial features (such as Turbo Options and Vaults), the legacy UI became a bottleneck in terms of usability, clarity, and feature discoverability. My role was to initiate and lead a strategic redesign that could support this growth, align stakeholder expectations, and deliver a more intuitive, scalable, and future-proof user experience.

Team Structure

2

Product
Managers

1

Business
& Growth

3

Developers

TL;DR - Rollover Feature Design

Problem

Users had to manually re-deposit into weekly markets. The legacy UI offered no clarity, no automation, and no visibility, hurting retention and liquidity.

My Role

Led end-to-end feature design process from community feedback synthesis, journey mapping, and sitemap planning to flow redesign, UI concepting, and dev collaboration.

Impact

Rollover-related deposits increased by ~28%
Reduced support friction & confusion
Team adopted more design-thinking-driven processes

One of the hardest design challenges I’ve tackled turned into one of the most scalable UX solutions on the product.

Methods Followed

Workshop -Product Map

We ran a full-team FigJam workshop with stakeholders across Product, Engineering, and Business to analyze the legacy UI flows. Pain points were clustered into themes like “Market Comparison,” “Portfolio Understanding,” and “Position Feedback.”

Netrographic Research

User insights from Discord and usability heuristics pointed out key issues such as unclear market details, repetitive actions, and discoverability problems, especially around epoch-based deposits.

Sitemap & Wireframes

We introduced a new flow for “Rollover,” solving manual repetition. Users could now commit to multiple epochs in advance, saving time while improving product stickiness.

Dev Handoff

Worked with another designer to build a design system that ensured scalable, consistent UI. Provided specs, component tokens, and detailed flows for engineers to implement easily.

Usability Validation

Tested prototypes with early adopters and received confirmation that users were able to understand and use the Rollover feature without additional instructions. The new market info display increased trust and confidence.

Context & Problem

#Validation
#Test

Context

The Rollover feature was designed to simplify the recurring process of opening positions in weekly DeFi markets.In Y2K, users could mint into two opposing sides of a prediction market each week (each “epoch” lasting 7 days)

Collateral: A bet that the asset will stay below a threshold (e.g., USDC stays under $0.9900)

Premium: A bet that the asset will exceed that same threshold

Here’s how it worked:If 10 users deposited $10 into Collateral, and 2 users deposited $10 into Premium, the total pool was $120.

If the Collateral condition held true, the $20 from the Premium side would be redistributed proportionally to the 10 users on the Collateral side (earning $2 each).

If Premium won, those 2 users would split the entire $100 from the Collateral side, earning $50 each.

The system worked, but it required manual interaction every single week, with no option to automate position renewal.

Problem

Despite the logic being sound, the user experience fell short, especially for more committed or long-term users

User side

Manual burden: Users had to remember to deposit again each week to stay in the market.

Missing context:
The legacy UI didn’t clearly show basic market-level information like previous epoch performance or upcoming thresholds.

Discoverability issues:
Users were unsure if their Rollover was successful or even triggered, there was no feedback or state-based guidance.

Business side

Rollover adoption meant more capital locked into the protocol ahead of time, improving TVL (Total Value Locked).

Encouraging users to set future epochs would increase predictability and liquidity.

Technical side

The legacy infrastructure was not built with automated logic or UI feedback layers, making it hard to communicate Rollover status across epochs.

Discovery & Insight

#Discovery

🔎 Community Signals

While exploring improvements for the Rollover experience, I collected direct feedback from active community members on Discord. These power users and early adopters regularly participated in weekly epochs and shared their pain points around both visibility and actionability:

Some representative quotes:

I’d like to see historical oracle price data—not just the current snapshot.”
“I want to compare yields across all markets before choosing which to roll into.”
“I wish I could see past epoch performance and payouts in one place.”
“Rollover, Queue, Edit, Delist—these controls should be together in one screen.”

The overall message was clear:
Community wanted more context, more control, and less fragmentation when managing recurring positions.

Journey Map Takeaways

As part of the earlier journey mapping workshop, we examined user paths related to Depositing and Rollover interactions. We found:

Missing visibility

Users lacked access to essential market history (TVL trends, depeg probabilities, ROI estimations).

Disconnected decision flow

The data users needed to make informed decisions was scattered or completely unavailable.

Action fragmentation

Users had to move between different screens to queue a position, set a rollover, or delist a market hurting usability.

Legacy UI Observations

Instead, users had to manually re-deposit every week through the same flow used for first-time deposits, leading to repetitive behavior and frustration.
While the feature technically existed, it failed to meet user needs. The legacy UI lacked both visual clarity and interaction support to guide users effectively.
Key information such as historical oracle data, market performance, rollover state, and comparison across epochs was either missing or buried across different screens.

From a visual and interaction design standpoint, there was no clear mental model for how to manage or automate future deposits — a critical usability gap for a feature meant to save time and reduce friction.

The Markets page was overly dense, lacking hierarchy and clarity.

No indicators or progress visuals helped users understand what stage their positions were in.

No way to “trust” that a Rollover was scheduled or completed—creating uncertainty and reducing engagement.

💡
Community were asking for clarity, control, and continuity over what they were already doing manually. This shifted the challenge from “adding a button” to designing a reliable, context-rich flow that simplifies and encourages participation in Rollover.

Solution

#Discovery

Re-Architecture

Designing the Rollover flow turned out to be one of the most complex design challenges I've faced. The goal was to strike a balance between flexibility and simplicity, all within the constratints of a dense, data-driven DeFi product.

We began by asking:

Where does the user expect to trigger Rollover?

We mapped entry points acroll the platform, from My Portfolio, to Market Details, to Strategy to ensure contextual accessibility. Then, we outlined the essential data and actions users needed to make decisions:

Why a Drawer Modal?

We considered building a full page, but opted for a modal structure instead.

Here is why?

Context preservation: User could easily go back to where they started
Reduced friction: Less navigation, more focus
Quicker interaction: Better suited for high-frequency users

We then built the flow step-by-step with modular components:

Rollover Modal: Split into "Your Original Position" and "Your Rollover Position"
Dynamic controls for inputs
Real-time calculation of ROI, Emissions, Relayer fee

Interface Enhancements

Visual feedback

Badge system added to indicate

Rolling Over

Queued Over

Won

Liquidated

Delisted

Modular Card Components
Post-rollover, we updated the My Positions view with rolled-over card indicators, showing upcoming epochs and expected performance at a glance.

Validation & Outcome

#Discovery

Adoption & Discoverability

Users immediately found the new Rollover modal intuitive, thanks to its inline placement within the portfolio view and consistent badge language across the app.

Entry points mapped during the design phase proved accurate users triggered Rollover primarily from Portfolio and Market Details, confirming the sitemap placement and access hierarchy worked.

Badge indicators (Rolling Over, Queued, etc.) were frequently mentioned by community members as helpful in understanding position states across epochs.

“I love that I can finally just queue my position and walk away, feels solid now!”
Community member on Discord

Product & My Impact

#Discovery

Adoption & Discoverability

Product Impact

The redesigned Rollover experience made it easier for users to stay engaged with Y2K’s prediction markets across multiple epochs—without needing to start over each week.

The new modal simplified a previously manual and error-prone flow

Data clarity (TVL, ROI, emissions) helped users make more informed decisions

Badge indicators and contextual UI reduced friction and increased confidence

As a result, weekly deposits via Rollover increased by ~28%, improving liquidity and user retention

My Impact

Beyond the feature itself, this project helped drive broader design maturity within the team:

Introduced structured UX methodologies such as journey mapping, sitemap planning, and heuristic evaluation

Facilitated cross-functional alignment through collaborative FigJam sessions

Brought design thinking into product planning, helping engineers and stakeholders better understand user needs

Established modular component-based thinking for UI decisions, still reused in other parts of the platform

Fostered a shared design language between design, product, and engineering