Carepool iOS — Human-centered design for seniors and people with disabilities
Designing an accessible ridesharing app for seniors and people with disabilities using AI-accelerated research and rapid prototyping. My Role: Led UX research and interaction design, focusing on accessibility standards (WCAG compliance). Leveraged AI tools to accelerate research synthesis and prototype iteration. Impact: Increased adoption and trust among underserved users by reducing barriers to mobility. Validated core features with users 40% faster through AI-assisted prototyping. Scale: Rolled out regionally with a growing user base of riders and caregivers.
Problem
Elderly and disabled users faced challenges with existing ride-sharing services like Uber and Lyft, which did not provide drivers trained in healthcare protocols, door-to-door assistance, or reassurance for users with mobility needs. The app needed to allow users to book rides independently without relying on caregivers, while ensuring safety and accessibility.
Solution
Key features included: • First-time booking simplicity: users could easily schedule a ride without confusion • Ride history: previous rides saved for repeat bookings, reducing friction for users who may struggle with memory or technology • Driver transparency: users could see the driver's profile, license plate, and estimated arrival time, providing trust and peace of mind The app followed WCAG accessibility guidelines to ensure usability for elderly and disabled users.
Research: Moving Fast with AI-Assisted Insights
Primary Research
Conducted user interviews with elderly and disabled users to understand their pain points and experiences with ride sharing. Used AI transcription and analysis tools to:
Auto-transcribe interview recordings in real-time
Extract key themes and pain points across multiple sessions
Identify patterns in user language and emotional responses 30% faster than manual coding
Competitive Analysis
Leveraged AI-powered competitive research tools to rapidly analyze existing ride-share services and their accessibility features. Used large language models to:
Synthesize accessibility reviews from app stores and forums
Identify feature gaps across 10+ competitors in hours vs. days
Generate comparison matrices highlighting opportunities for differentiation
Key Findings
Users wanted drivers who understood their needs and could provide door-to-door assistance, including handling wheelchairs
Users needed reassurance that drivers had healthcare experience and could be trusted
Ease-of-use and simplicity were critical for first-time users
AI revealed an unexpected insight: User interview analysis surfaced recurring anxiety about communication during pickups, leading to the addition of simple in-app messaging templates
Design Process: Rapid Prototyping to Get in Front of Users
Week 1: AI-Accelerated Ideation
Used AI image generation tools to quickly explore visual directions for accessibility-focused UI
Generated multiple layout variations for booking flows using AI design assistants
Created accessibility-compliant color palettes in minutes using AI tools that automatically checked WCAG contrast ratios
Week 2: Fast Wireframing & Early Validation
Built initial wireframes and reviewed with CEO, CTO, PM, and developers
Used AI to generate multiple variations of critical screens (booking flow, driver profile, ride history)
Got in front of users early: Tested low-fidelity prototypes with 5 users by end of Week 2 - typically a Week 4-5 milestone
Week 3-4: Interactive Prototypes & Iteration
Developed interactive prototypes using AI-assisted design tools that auto-generated responsive layouts
Conducted rapid usability testing with users, gathering feedback on layout, readability, and ease of booking
Accelerated iteration: Used AI to quickly generate alternative solutions for problem areas identified in testing
Real-time synthesis: AI tools helped process user feedback recordings overnight, enabling same-week design iterations
Continuous Testing & Refinement
By leveraging AI for the heavy lifting of transcription, synthesis, and variation generation, the team conducted 3 rounds of user testing in the time normally allocated for 1-2 rounds, ensuring accessibility and simplicity through data-driven iterations.
Collaboration
Daily standups and weekly demos kept the small team aligned. AI tools enhanced collaboration by:
Auto-generating design documentation and specs for developers
Creating accessibility audit reports that flagged potential WCAG violations before handoff
Maintaining a smooth, accelerated handoff from design to development with less manual documentation
Results & Impact
User Reception
Upon launch, users who hadn't seen prototypes were excited and grateful for a service they could trust. Users expressed relief in knowing the driver had healthcare experience, could assist with mobility needs, and that the booking process was simple and reliable.
Speed to Market
40% faster prototype iteration through AI-assisted design tools
3x more user testing rounds in the same timeframe, leading to higher confidence in design decisions
Earlier user validation allowed the team to pivot on key features before significant development investment
Qualitative Impact
Qualitative feedback highlighted the app's impact on independence, trust, and confidence in getting to appointments and daily activities. The in-app messaging templates, discovered through AI-assisted research analysis, became one of the most appreciated features for users with speech or hearing challenges.
Reflection
What I Learned
AI as a research accelerator: Gained deep empathy for elderly and disabled users while using AI to handle time-consuming synthesis tasks, allowing more time for actual user conversations
Prototyping velocity matters: Getting rough prototypes in front of users by Week 2 uncovered critical usability issues that would have been expensive to fix post-development
Trust through transparency: Learned how to design for trust in the ride-share industry, with AI competitive analysis revealing best practices across adjacent healthcare transportation services
Future Opportunities
With the AI-enhanced workflow established, future iterations could:
Use AI sentiment analysis on ongoing user feedback to proactively identify friction points
Leverage predictive models to personalize the experience for users with different accessibility needs
Conduct continuous, lightweight user testing using AI-moderated research tools to maintain user-centricity at scale
The MVP successfully demonstrated the value of a direct-to-user ride-sharing solution, and the AI-accelerated process proved that accessible, human-centered design doesn't have to move slowly, it just needs to move thoughtfully.
