Good care
is a shared
thing.
Appointments, next steps and the small things that matter—all in one kind, considered place for you and the people who care about you.
Explore a care plan ↗A little progress
goes a long way, Alex.
One day at a time. Here’s what’s coming up.
2 of 3 next steps completed
“You don’t have to remember everything on your own.”
Keep your plans together.
Keep decisions yours.
Careloom organizes care tasks into a calm timeline, makes questions easy to prepare and lets the user decide who to involve. The concept avoids diagnosis, treatment recommendations or automatic sharing.
Careloom / Care navigation
A post-appointment companion that helps people organize follow-up tasks, visits and questions in one approachable view.
The problem
After appointments, patients often manage instructions, scheduling and administrative tasks across multiple channels, increasing confusion about the next action.
Product strategy
Soft, calm language, progressive disclosure and high accessibility. The product emphasizes an editable to-do list rather than a potentially unsafe diagnostic assistant.
Design decisions
Organic, calm, rounded / ivory and sage. Use a single next-action card, accessible hierarchy and editable non-clinical checklists.
UX research approach
Synthetic research: 8 fictional patient-caregiver profiles, 10 modeled navigation tasks. This is a fictional scenario: no real interviews or measured outcomes are claimed.
Accessibility & responsive design
Keyboard-operable controls, visible focus, readable content and responsive reflow support this concept. Reduced-motion preferences remove animated scrolling. Physical-device and assistive-technology validation remain future work.
AI-assisted workflow
Prioritize readability, screen-reader order, large touch targets and consent; mock reminders with no patient identifiers. AI-assisted implementation is used for prototyping and iteration; product recommendations here are deterministic local simulations.
Proposed technical architecture
React, TypeScript, accessible component library, FHIR-compatible integration layer, encrypted datastore and role-based access control. Entirely synthetic, non-patient example tasks; no medical records. This architecture is a proposal, not a deployed backend.
Limitations & lessons
Clinical uncertainty leads to contact-your-care-team messaging. Never generate diagnoses, medication changes or clinical recommendations. The next step is real user research, production data validation and accessibility testing before any launch.
Information architecture & user flow
- User selects a generic follow-up task
- Agent organizes the non-clinical next-step checklist
- User confirms or edits suggested reminders
- Human care team remains source of clinical instructions
Core views: My next steps overview → Task clarification and questions → Visit scheduling handoff