AI Health: From Records to Action

Connect for Patients (C4P)

COMPANY
OVERVIEW

A 0→1 consumer health app that turns a lifetime of scattered medical records into one clear signal patients can act on.

YEAR

YEAR

2025-2026

2025-2026

ROLE

Solo Product Designer: product logic, end-to-end UX, and the design system

TEAM

TEAM

CPO (product direction), Engineering

TIMELINE

TIMELINE

0→1 to launch in 9 months

PLATFORM

PLATFORM

iOS, Android, web app

One source of truth for a patient's whole health

The C4P patient app gives people their entire health in one place: every record from every provider (FHIR, HIPAA-authorized), cleansed and verified by C4P's AI, Orion, then made understandable and actionable. That same connected record becomes one complete record for their providers, too.

I designed it end to end as the solo product designer, turning a high-level vision into a real, shippable product: the logic, the flows, and every screen.

Patients don't need more numbers, they need a read

A patient's health data lives scattered across providers, labs, wearables, and pharmacies. Even once it's gathered, it's a wall of numbers that means nothing to a normal person. Patients don't need more data, they need to know "Am I okay, and what should I do?"

And on the business side, raw data alone wasn't a strong enough reason for a patient to connect. The app had to give them something they'd want for themselves.

One system, so one designer could move like a team

I built and maintain the design system from 0 as the single source of truth for a fast-moving product. To keep design, AI, and code in sync, I carried it across three surfaces:

  • Figma component library for design

  • Claude-based design system for AI-assisted work

  • Front-end implementation for engineering

Three deliveries, one source of truth, so a solo designer could move at the speed of a team.

Deep Dive

The Vitality evaluation system

The logic behind screens

Three roll-up layers, one glance

The challenge wasn't the screens. It was turning any patient's raw FHIR data into one honest read that adapts to them: meaningful whether they're healthy or chronically ill.

The real design work was the framework that decides what a patient sees and why, turning clinical reality into something any users can feel in one glance and act on.

How the product thinks

AI as the backbone, not a chatbot

I designed Orion AI as the product’s backbone, not a standalone chatbot. Its intelligence appears across the app as structured, trustworthy guidance.

For Vitality, I created the health evaluation system and calculation rules. Orion interprets baseline data, scores each goal, and rolls the results into three categories and one overall Vitality read. AI is present at every layer, not just one step.

Key decision & tradeoff

From tracker to framework

My first design was a heavier tracking system that logged every to-do under every goal, visualized like a tracker. Too complex, not intuitive, and near-impossible to build from scratch. I cut it. The lesson: from 0, you need a clear, just-right skeleton the product can grow into — not a feature-complete maze.

The outcome

Impact & Reflection

9mo

Concept to launch

2 weeks

To first paying customers

300+

Users onboarded in 2mo

I expected to design screens. I ended up designing how the product thinks, decides, and grows, and doing it for users and data that didn't exist yet: building a frame and letting reality fill it in, making calls that would still hold when the product went from thin to crowded. With no PM to hand the hard questions to, this is where I became a product designer.

A laptop with a h32 word on it.