AIDSVU & HEPVU

Interactive Map & Location Profiles

The Problem

Emory University came to us with an existing AIDSvu platform that mapped HIV prevalence across the United States. The platform worked, but it had hit a wall. Users were constantly contacting the team asking for data exports for specific areas. The research team wanted to add more data and more indicators to tell a fuller story about HIV in America. But adding more to the interface meant more complexity and more overwhelm.

The challenge was to redesign the platform so it could handle significantly more indicators, more comparison options, and more user control without becoming cluttered or unusable.

My Role

I was the sole designer on a cross-functional team working alongside a PM, analyst, developers, and researchers from Emory. My job was to understand what data the research team wanted to expose, how users actually wanted to interact with it, and then design an interface that felt simple on the surface but could go as deep as a researcher needed.

Key Design Decisions

1. Progressive disclosure to reveal complexity gradually

Instead of showing all indicators at once, I designed the interface around progressive disclosure. Users start with a clean map showing a single metric like HIV prevalence. From there, they can choose to layer on additional metrics, compare two maps side by side, or dig into specific demographics. The interface never feels crowded because complexity is revealed only when users ask for it.

2. Side-by-side map comparison

One of the most requested features was comparing metrics against each other. I designed a split-screen view where users could compare two maps simultaneously: their home state against the national average, or new diagnoses against prevalence, or any combination they needed. Both maps stay locked to the same zoom level and region so the comparison is intuitive and immediate.

3. Granular filtering without breaking the interface

The redesign needed to support filtering by age, gender, transmission type, and more. I designed a collapsible filter panel that revealed options progressively. Users could start simple (show me one age group) and add complexity as they needed it. Every filter change updated both the map and the data in real time, so the feedback loop was instant.

4. Location reports that tell a story

When users clicked on a location or state, the old platform showed raw numbers. In the redesign, I turned those into narrative reports. Key metrics are compared side by side (this state vs. national average, this year vs. last year, this demographic vs. overall). The report gives context and tells a story rather than just displaying data. Users can read it on screen or export a PDF that looks professional enough to bring to a health department meeting.

5. One visual system across multiple metrics

The platform now supports prevalence, new diagnoses, opioid-related indicators, and more. I designed a unified visual language across all of them: consistent color palettes tested for accessibility, consistent chart types, consistent interaction patterns. Adding a new metric to the platform doesn't require a redesign. The system scales.

What Changed

The original AIDSvu gave users one map view. The redesigned version gives them control: which metrics to show, which areas to compare, which demographics to filter, and how deep to go. Users could spend 30 seconds getting a quick answer or two hours doing deep epidemiological analysis on the same platform.

The redesign also led directly to HepVu, a companion platform for Hepatitis C data built on the same design system and interaction patterns. The system was designed to scale, and it did.

Outcome

• Award-winning platform used by health departments, researchers, and advocacy organizations across the U.S.
• Redesign directly led to HepVu, a companion platform built on the same design system
• Continues to evolve with annual updates based on new epidemiological data

Environmental Defense Fund

The U.S. Climate Vulnerability Index

The Problem

The Environmental Defense Fund had created a climate vulnerability index with rigorous data across 70,000+ U.S. census tracts and 184 indicators. The data was powerful, but it lived in spreadsheets and research reports. What was missing was a way for everyday people to find their own community, understand what makes it vulnerable, and take that knowledge to their government to advocate for change.

The challenge wasn't just making the data accessible. It was designing a platform where someone could find their neighborhood, understand their specific vulnerabilities, whether that's living near a power plant or facing extreme heat risk, and walk away with a concrete, exportable report to bring to local government.

My Role

I was the sole designer on a cross-functional team alongside a PM, developers, and a researcher/analyst from EDF. I owned the design end to end: information architecture, interaction design, high-fidelity UI, the project's design system, and accessibility. Day to day, I worked closely with the research team to understand the data, with developers on technical constraints, and with the PM to align priorities in regular sprints.

Key Design Decisions

1. Map as the entry point, reports as the action

The map is the hero because geography is how people think about their own vulnerability. But the map alone wasn't enough. Users needed to go from "I can see my area is red" to "here's exactly why, and here's what I can do about it." I designed two connected views: the map for exploration (zoom to your area, see layers of risk) and the location report for understanding (click a spot, read a narrative explanation, export to PDF for advocacy).

2. 184 indicators without overwhelming

184 indicators is a nightmare for any UI. Working with the research team, I designed a layered hierarchy: start with an overall vulnerability score visible on the map, then let users reveal thematic categories (health, environment, infrastructure, social vulnerability), and drill into individual indicators only when they want to. A casual user could understand their risk in 30 seconds. A researcher could spend an hour going deep. Same interface.

3. Redundant visual encoding for accessibility

Maps are tricky for accessibility. I designed every data layer with color plus pattern plus labels so users with color vision deficiency could still read the data. All interactions worked via keyboard. Charts and filters shipped with semantic markup so screen reader users got the same depth as sighted users. WCAG 2.1 compliance was non-negotiable from day one.

4. Export as a core feature, not an afterthought

Users came to this platform to advocate. That meant reports had to be export-friendly from the start. I designed the location report layout to look clean and credible in both screen and print: headers, charts, key statistics, and a plain-language summary of vulnerabilities, all rendering correctly when exported to PDF.

5. Responsive without losing depth

Someone checking their neighborhood's vulnerability on their phone needed the same clarity as someone on a desktop. I designed mobile interactions for every component: touch-friendly map controls, modal-based filters, and reports that stayed readable and exportable on small screens.

Outcome

• Anthem Award winner, Best Use of Data
• Featured in news coverage and academic research
• Used by community organizations, policymakers, and researchers across the U.S.
• Continues to evolve through ongoing partnership with EDF

Economic Policy Institute

State of Working America Data Library

The Problem

The Economic Policy Institute had years of rigorous economic data: wage trends, inequality metrics, demographics, state-by-state comparisons. But their existing platform was purely transactional. Users would come, select the data they wanted, and download it as an Excel sheet or CSV. There was no way to explore, no way to compare, no way to discover patterns. You had to know exactly what you wanted before you arrived.

The challenge was to transform a download tool into a platform where journalists, policymakers, researchers, and curious citizens could actually explore economic inequality and find answers they didn't know they were looking for.

My Role

I was the sole designer on a cross-functional team alongside a PM, developers, and an economist from EPI. I worked closely with the research team to understand what stories the data could tell, then designed an interface that made those stories discoverable without requiring users to be data analysts themselves.

Key Design Decisions

1. From export tool to exploration platform

The old site was a form: select your dataset, select your format, download. The new platform is a library. Users browse available datasets, see what each one measures, and click to explore. Each dataset opens into a visualization where users can immediately see the data in context. Downloading is still there, but it's no longer the only thing you can do.

2. Flexible comparison as the core interaction

The most powerful feature is comparison. I designed the interface so users could ask any "versus" question the data supported: California versus New York wages, wages by age group, wages by gender, median income by state. Each comparison generates a new view. Users can layer comparisons without ever leaving the interface.

3. Multiple views for the same data

Some users want to see one metric across all states on a map. Others want to see trends over time in a chart. Others want raw numbers in a table. I designed flexible views so users could switch between map, chart, comparison, and table views. The same underlying data, different lenses depending on what the user needs to understand. You could look at average wages by state on a map, then switch to a line chart showing how wages have changed over the past decade.

4. Progressive disclosure so users don't get lost

EPI tracks dozens of metrics across multiple dimensions. Instead of overwhelming users with all options at once, I designed the interface to reveal filters progressively. Start with a simple metric like median wage, then add filters by age, gender, or location only as you need them. Each filter narrows the data without cluttering the screen.

5. Export is still powerful, it's just not the only goal

The old platform's entire purpose was export. I kept that functionality intact, but it's now one of many actions. Users can explore for ten minutes, find something interesting, and export just that slice of data. Or share a link to their specific view. Or screenshot a chart for an article. The data became useful whether or not anyone downloads a spreadsheet.

What Changed

Journalists used the platform to find story angles they wouldn't have discovered by downloading CSVs. Policy advocates used it to build visual cases for wage reform. Teachers used it to show students real economic data in a way that made sense. The platform became a research and discovery tool, not just a data warehouse.

Outcome

• Award-winning platform used by journalists, policymakers, researchers, and educators
• Data from the library cited in news coverage, policy briefs, and academic research
• Transformed EPI's data from a download service into a public resource people actually explore