Navora
The Problem
Discharge isn't the end of care.
Hospitals give patients instructions. But once they leave, the work of actually completing that care falls on them.
A new mother may need to find a pelvic-floor therapist, schedule an appointment, navigate insurance, arrange transportation, manage medications, and keep track of follow-ups — often while recovering herself.
The healthcare system gives patients a plan. Navora helps execute it.
What I Learned
The problem isn't information. It's execution.
Through conversations with patients and hospital staff, I kept seeing the same gap: the instructions already exist. What's missing is the coordination layer between “here's what you need to do” and “it's done.”
Core Design Principles
Reduce cognitive load: Don't give someone recovering from a major medical event another dashboard to manage.
Turn plans into actions: Every recommendation should lead toward a concrete next step.
Meet people where they are: Navora should feel lightweight and accessible, not like another healthcare system to learn.
Use AI where it removes work: AI handles the coordination and administrative complexity so the patient doesn't have to.
Exploration
Finding the right shape for Navora.
Before committing to the final experience, I explored different visual directions, interaction patterns, and ways AI could make postpartum support feel more approachable and human.
01 — Visual exploration
Different directions for Navora's visual identity, tone, color, and brand language.
02 — Interaction exploration
Exploring how a text-first AI experience could feel natural rather than like another healthcare dashboard.
03 — Product exploration
Testing different ways to turn discharge instructions into actionable next steps.
Inspiration
Borrowing from products that already understand the user.
I looked at how Apple and ChatGPT approach two things Navora needed to get right: making complex information feel simple, and making an AI interaction feel natural.
Apple Health / Fitness
I was inspired by how Apple turns large amounts of health data into simple, understandable signals. The goal wasn't to recreate a health dashboard, but to make Navora's recommendations feel equally clear and actionable.ChatGPT
ChatGPT became an important reference for the interaction model. Rather than forcing patients through complicated forms and workflows, I explored a conversational interface where they could simply tell Navora what they needed and let the system handle the complexity.
From Figma to Working Product
I used Claude Code alongside Figma to rapidly turn concepts into functional product experiences. This let me test real workflows, iterate quickly, and make design decisions based on how the product actually behaved — not just how it looked.
Key Results
Built & launched solo: Took Navora from concept to working product across strategy, UX, design, and engineering.
Validated the problem: Interviewed 120+ moms and 25+ hospital staff across five hospitals to understand where care breaks down after discharge.
Early adoption: 50 moms have onboarded, with two hospitals in pilot conversations.
Built clinical credibility: Brought on Dr. Emma Ramsey as a clinical advisor, with additional clinicians engaged around the product.
What Changed & What I Learned
The patient doesn't need more information: Discharge instructions already contain plenty of information. The real gap is turning those instructions into actions someone can actually complete.
Good healthcare UX reduces cognitive load: Recovery is already demanding. Every interaction should remove work from the patient, not create another system they have to learn.
AI should disappear into the experience: The most valuable role for AI isn't being the product. It's quietly handling the complexity behind the scenes so the experience can stay simple.
Designing for trust changes everything: When you're working with someone's health, the product has to feel clear, predictable, and human before it can ask someone to rely on it.
