TL;DR

What: Designed Relay, a concern routing system for dialysis clinics. Captures patient concerns at bedside in under 10 seconds, routes them to the right clinician, and tracks each one to resolution.

Why: Up to 60% of patient-reported concerns vanish between being spoken and being documented. In chronic care where patients return 3x/week, that compounds into eroded trust and missed clinical signals.

| Why: Up to 60% of patient concerns in dialysis clinics disappear between being spoken and being acted on. Because there's no reliable path from the person who hears it to the person who can resolve it.

Result: Selected as 3/80 externs to present to Stanford clinicians.

| Outcome: Selected as 3/80 externs to present to Stanford clinicians. Projected to raise concern capture rate from 60% to 80%.

Result: Selected as 3/80 externs to present to Stanford clinicians.

Result: Selected as 3/80 externs to present to Stanford clinicians.

| Result: Built a lifecyle simulator MVP & selected as 3/80 externs to present to Stanford clinicians.

Role: Owned end-to-end design process.

| My Role: Solo designer. Owned the full process, from diagnosis to concept to pre-validation across 4 clinical scenarios.

Role: Owned end-to-end design process.

MVP: An interactive prototype walking through Relay's full concern lifecycle, from bedside capture to clinical resolution. This also includes any edge cases.

| My Role: Solo designer. Owned the full process, from diagnosis to concept to pre-validation across 4 clinical scenarios.

MVP: An interactive prototype walking through Relay's full concern lifecycle, from bedside capture to clinical resolution. This also includes any edge cases.

View MVP

THE CLINIC'S STARTING PROBLEM

Upto 60% patient concerns disappear during dialysis sessions.

A patient mentions leg cramps during their session. Here's what happens next.

Patient Care Technician

Hears the concern.
But can't make clinical decisions pulled to another chair.

Registered Nurse

Could assess it.
But wasn't there when the patient spoke. Never receives it.

Medical Doctor

Could act on it.
But only sees what gets escalated. Mostly not around.

The problem isn’t recording verbal concerns, it’s routing them.
They’re captured, but don’t reach the person who can resolve them on time.

WHY THIS IS HARD TO FIX

Concerns surface in seconds, but resolution requires coordination across roles.

Three insights that make this a design problem.

THE DESIGN CHALLENGE

How might we design a reliable path between a patient voicing a concern and the clinic acting on it?

The person who hears it can’t triage it. The person who can triage it may not be there. The person who can act on it only sees what gets escalated. The system has to connect all three, without adding burden to any of them.

CONSTRAINTS TO PRINCIPLES

Each constraint became a design principle.

01 Capture before it's forgotten

Insight: PCTs are moving between patients. Capture needs to be fast.

02 Ownership by assignment

Insight: Without explicit assignment, tasks fall between roles.

03 Time Bound Responses

Insight: Delays impact patient trust. Response time is crucial.

04 Context survives handoffs

Insight: Handoffs lose 30% of context. History must travel with the concern.

MY DESIGN PROCESS

Designing for Clinic's Ground Reality, with AI

My AI-Augmented Design process has its roots in the classic double diamond. I front-loaded the process around understanding the clinic roles & context, and used AI to go deeper.

Product Brief

EXPLORATIONS

Three directions rejected. One tradeoff accepted.

I explored three approaches, each solved part of the problem but failed on the rest. Relay came out to be the winner with two tradeoffs: complex build & EMR (electronic medical records) integration.

Kanban Board

Decision Support Tool

EMR Checkpoint

THE SOLUTION

Relay: A concern routing system that guarantees every patient concern is captured, owned, and resolved.

Concerns need a trajectory not better documentation. Relay moves a concern from the moment a patient speaks, through triage and treatment, to documented closure.

The workflow maps every concern level, from PCT capture to RN escalation to MD intervention, with clear decision points and role handoffs at each step.

PHASE 1: CAPTURE

A patient says something.

The technician has 10 seconds before they forget or get pulled to another chair.

A patient says something. The technician has 10 seconds before they forget or get pulled to another chair.

This screen exists to move words from memory into the system fast.

AI Iterations

AI Iterations

AI Iterations

AI Iterations

Human-Owned Designs

Human-Owned Designs

PHASE 2: CLASSIFY

The Nurse just received a concern from the Technician.

Now they decide: How urgent is this? Who should handle it?

The Nurse just received a concern from the Technician. Now they decide: How urgent is this? Who should handle it?

Three levels of concerns: comfort, non-urgent clinical, urgent to keep classification fast.

AI Iterations

Human-Owned Designs

Human-Owned Designs

Human-Owned Designs

The nurse makes a clinical judgment. The system handles routing, ownership, notifications, and deadlines.

| This avoids two common failures.

Over-documenting simple issues and vague escalation when urgency is real.

PHASE 3: ESCALATE (NURSE TO DOCTOR)

The leg cramps turn out to be vascular.
Nurse classifies the concern as urgent.

The doctor gets an alert immediately.

Doctors aren't at the bedside. They need context fast. The nurse's note enables a decision. If they can't act, ownership transfers automatically, no concern gets stuck.

The doctor accepted.

Now the nurse & doctor are coordinating
care remotely while the nurse works at the bedside.

The doctor accepted. Now they're coordinating care remotely while the nurse works at the bedside.

Both see the same patient. Both stay in sync.

Doctor's View

Nurse's View

During an emergency, the nurse's hands are busy. A single status selector (Stabilizing → Responding → Not Responding → Stable) takes one tap. The doctor sees it instantly.

PHASE 4: CLOSE

The patient is stable.

But the nurse can't close this alone.
L3 concerns require doctor sign-off.

The patient is stable. But the nurse can't close this alone. L3 concerns require doctor sign-off.

Doctor sees the nurse's note upfront. They can request more info before accepting. The concern closes & is marked as resolved or handed off after the doctor approves it.

| Nurses can’t close high-severity concerns due to regulation.

They document; doctors review and approve, ensuring accountability and fast resolution.

DOCTOR FEEDBACK

I was selected as 1 of the 3 externs from a batch of 80 to present my solution to get real feedback.

"The tool must reduce mental load in the moment and feel “native” to clinic workflow, not like extra admin"

~ Kelly Chen, Nephrology Nurse Practitioner

PRE-VALIDATION

Simulated walkthroughs before real clinic testing.

I used a synthetic research process to pre-validate the concept to reduce risk before a live pilot.

Walked 6 scenarios through 4 roles

Each scenario imposed a real constraint: peak-hour capture, escalation with the RN absent, ambiguous symptom urgency, shift-change handoff, reassignments, and recurrent concerns.

The workflow kept breaking in the same place

The decision tree outlines the actions each person needs to do. But across multiple scenarios, one role had no way to see whether any of it was actually happening.

The charge nurse was accountable, but blind

She manages the floor, reassigns staff, and triages priorities. But she had no view of which concerns were captured, which were stuck in escalation, and which were missed entirely.

Charge Nurse (CN)

| Owns the floor, but can't see what's happening on it

Knows staffing levels, shift timeline, escalation status
Acts in the full 8-hour shift window
Can't assess patients or be at every station

Charge Nurse (CN)

| Owns the floor, but can't see what's happening on it

Knows staffing levels, shift timeline, escalation status
Acts in the full 8-hour shift window
Can't assess patients or be at every station

Charge Nurse (CN)

| Owns the floor, but can't see what's happening on it

Knows staffing levels, shift timeline, escalation status
Acts in the full 8-hour shift window
Can't assess patients or be at every station

That gap became the dashboard

The charge nurse needed a real-time view of what was happening across her floor, concern status, ownership, documentation gaps, and escalation state at every station.

PROJECTED IMPACT

These outcomes guided every design decision and would be the first metrics I'd track in a pilot.

REFLECTION

Failures & Learnings from Designing for Healthcare

I kept designing
for ideal conditions.

I kept designing for
ideal conditions.

Learning to design for
the worst moment, not the
best one, changed how I approach every screen now.

Learning to design for the worst moment, not the
best one, changed how I approach every screen now.

I underestimated
how much info dies
in transitions.

I underestimated how much
info dies in transitions.

I thought capture was
enough. I missed how vital
handoffs are in healthcare.

I thought capture was enough. I missed how vital
handoffs are in healthcare.

I confused visibility
with usefulness.

I confused visibility with usefulness.

My instinct was to show
more. Learning to surface
only what needs attention
reshaped how I design.

My instinct was to show more. Learning to surface
only what needs attention reshaped how I design.

Principles I've added to my toolkit:

Designing for reliability is different from designing for the ideal behaviour.

AI is most valuable when it reduces friction not for making decisions.

I kept designing for
ideal conditions.

Learning to design for the worst moment, not the best one, changed how I approach every screen now.

I underestimated how much
info dies in transitions.

I thought capture was enough. I missed how vital handoffs are in healthcare.

I confused visibility with usefulness.

My instinct was to show more. Learning to surface only what needs attention reshaped how I design.

I kept designing for
ideal conditions.

Learning to design for the worst moment, not the best one, changed how I approach every screen now.

I underestimated how much
info dies in transitions.

I thought capture was enough. I missed how vital handoffs are in healthcare.

I confused visibility with usefulness.

My instinct was to show more. Learning to surface only what needs attention reshaped how I design.

I kept designing for
ideal conditions.

Learning to design for the worst moment, not the best one, changed how I approach every screen now.

I underestimated how much info dies in transitions.

I thought capture was enough. I missed how vital handoffs are in healthcare.

I confused visibility with usefulness.

My instinct was to show more. Learning to surface only what needs attention reshaped how I design.

I kept designing for
ideal conditions.

Learning to design for the worst moment, not the best one, changed how I approach every screen now.

I underestimated how much info dies in transitions.

I thought capture was enough. I missed how vital handoffs are in healthcare.

I confused visibility with usefulness.

My instinct was to show more. Learning to surface only what needs attention reshaped how I design.

Routing Patient Concerns Before They Disappear

Routing Patient Concerns Before They Disappear

Upto 60% patient concerns disappear during dialysis sessions.

Upto 60% patient concerns disappear during dialysis sessions.

Patient Care Technician

Patient Care Technician

Registered Nurse

Registered Nurse

Medical Doctor

Medical Doctor

Concerns surface in seconds, but resolution requires coordination across roles.

Concerns surface in seconds, but resolution requires coordination across roles.

How might we design a reliable path between a patient voicing a concern and the clinic acting on it?

How might we design a reliable path between a patient voicing a concern and the clinic acting on it?

Each constraint became a design principle.

Each constraint became a design principle.

01 Capture before it's forgotten

01 Capture before it's forgotten

02 Ownership by assignment

02 Ownership by assignment

03 Time Bound Responses

03 Time Bound Responses

04 Context survives handoffs

04 Context survives handoffs

Designing for Clinic's Ground Reality, with AI

Designing for Clinic's Ground Reality, with AI

Three directions rejected. One tradeoff accepted.

Three directions rejected. One tradeoff accepted.

Relay: A concern routing system that guarantees every patient concern is captured, owned, and resolved.

Relay: A concern routing system that guarantees every patient concern is captured, owned, and resolved.

The workflow maps every concern level, from PCT capture to RN escalation to MD intervention, with clear decision points and role handoffs at each step.

The workflow maps every concern level, from PCT capture to RN escalation to MD intervention, with clear decision points and role handoffs at each step.

A patient says something. The technician has 10 seconds before they forget or get pulled to another chair.

A patient says something. The technician has 10 seconds before they forget or get pulled to another chair.

A patient says something. The technician has 10 seconds before they forget or get pulled to another chair.

The Nurse just received a concern from the Technician. Now they decide: How urgent is this? Who should handle it?

The Nurse just received a concern from the Technician. Now they decide: How urgent is this? Who should handle it?

The Nurse just received a concern from the Technician. Now they decide: How urgent is this? Who should handle it?

| This avoids two common failures.

| This avoids two common failures.

The leg cramps turn out to be vascular. Nurse classifies the concern as urgent.The doctor gets an alert immediately.

The leg cramps turn out to be vascular. Nurse classifies the concern as urgent.The doctor gets an alert immediately.

The doctor accepted. Now the nurse & doctor are coordinating care remotely while the nurse works at the bedside.

The doctor accepted. Now they're coordinating care remotely while the nurse works at the bedside.

The doctor accepted. Now they're coordinating care remotely while the nurse works at the bedside.

The patient is stable. But the nurse can't close this alone. L3 concerns require doctor sign-off.

The patient is stable. But the nurse can't close this alone. L3 concerns require doctor sign-off.

The patient is stable. But the nurse can't close this alone. L3 concerns require doctor sign-off.

| Nurses can’t close high-severity concerns due to regulation.

| Nurses can’t close high-severity concerns due to regulation.

"The tool must reduce mental load in the moment and feel “native” to clinic workflow, not like extra admin"

"The tool must reduce mental load in the moment and feel “native” to clinic workflow, not like extra admin"

Simulated walkthroughs before real clinic testing.

Simulated walkthroughs before real clinic testing.

Walked 6 scenarios through 4 roles

Walked 6 scenarios through 4 roles

The workflow kept breaking in the same place

The workflow kept breaking in the same place

The charge nurse was accountable, but blind

The charge nurse was accountable, but blind

Charge Nurse (CN)

Charge Nurse (CN)

| Owns the floor, but can't see what's happening on it

| Owns the floor, but can't see what's happening on it

Charge Nurse (CN)

Charge Nurse (CN)

That gap became the dashboard

That gap became the dashboard

These outcomes guided every design decision and would be the first metrics I'd track in a pilot.

These outcomes guided every design decision and would be the first metrics I'd track in a pilot.

Failures & Learnings from Designing for Healthcare

I kept designing for ideal conditions.

I kept designing for ideal conditions.

I underestimated how much info dies in transitions.

I underestimated how much info dies in transitions.

I confused visibility with usefulness.

I confused visibility with usefulness.

Principles I've added to my toolkit:

I kept designing for ideal conditions.

I underestimated how much info dies in transitions.

Routing Patient Concerns
Before They Disappear

Routing Patient Concerns
Before They Disappear

A patient says something.

The technician has 10 seconds before they forget or get pulled to another chair.

A patient says something.

The technician has 10 seconds before they forget or get pulled to another chair.

The Nurse just received a concern from the Technician.

Now they decide: How urgent is this? Who should handle it?

The leg cramps turn out to be vascular.
Nurse classifies the concern as urgent.

The doctor gets an alert immediately.

The leg cramps turn out to be vascular.
Nurse classifies the concern as urgent.

The doctor gets an alert immediately.

The doctor accepted.

Now the nurse & doctor are coordinating
care remotely while the nurse works at the bedside.

The patient is stable.

But the nurse can't close this alone.
L3 concerns require doctor sign-off.

I kept designing
for ideal conditions.

I kept designing for
ideal conditions.

I underestimated
how much info dies
in transitions.

I underestimated how much
info dies in transitions.

I confused visibility
with usefulness.

I kept designing for
ideal conditions.

I underestimated how much
info dies in transitions.

I kept designing for
ideal conditions.

I underestimated how much
info dies in transitions.

I kept designing for
ideal conditions.

I kept designing for
ideal conditions.