An AI-powered mental health companion that nurtures wellness through journaling, therapy, and personalized insights
My role
Lead Product designer
Impact I brought
Lead Product designer
Team
1 Lead Designer(Me)
1 CTO
1 Founding Designer
2 Supporting Designers
Tools
Figma, Miro, After effect
Platform
Mobile App (iOS, Android)
Duration
3 months
Problem
Most wellness apps speak. Few listen.
But our target users—Gen Z—needed something deeper.
They wanted…
Empathy, not efficiency.
Continuity, not one-off check-ins.
Emotional safety, not productivity hacks.
From early user interviews and diary studies, I uncovered a set of emotional failures:
Challenge #1
Feedback felt scripted, not human
Responses often sounded like templated coaching tips, lacking emotional depth or relevance.
Challenge #2
No memory or continuity over time
Users noticed the AI didn’t acknowledge prior input or build relational consistency.
Then I reframed the problem space…
Instead of building an AI therapist, what if I built a daily emotional companion? Not a tool for curing, but for caring. This reframing helped me define a product vision grounded in:
Product vision #1
Emotionally adaptive interaction
Product vision #2
Lightweight, consistent rituals
Product vision #3
User-driven tone and pace
Outcomes & Impact
A snapshot of the tangible results—from user behavior shifts to system-level wins.
Impact #1
Feedback felt scripted, not human
Responses often sounded like templated coaching tips, lacking emotional depth or relevance.
Impact #2
No memory or continuity over time
Users noticed the AI didn’t acknowledge prior input or build relational consistency.
Impact #3
Users dropped off when emotional effort outweighed perceived value
The interface required more emotional labor than the support it returned, especially after the first few days.
Solution
Introduce LePal.ai
— a wellness app that speaks with you, not at you.
Solution #1
🧠 My Journal
What it is:
A mood-aware journaling flow that adjusts prompts based on your emotional history.
Why it matters:
Because some days, you're ready to reflect. Other days, you just need a soft landing.
Solution #2
🔮 Crystal Ball
What it is:
A once-a-day micro-interaction that poses a reflective question—like a fortune cookie, but smarter.
Why it matters:
Because sometimes, the best nudge is the one you didn't know you needed.
Solution #3
🌍 Therapy Planet
What it is:
An AI conversation space where you can pick a topic—like burnout or love—and just talk.
Why it matters:
Because naming the problem is hard. We let you choose a door and take it from there.
Process
From Insight to Interaction
As the product designer on the team, I led the entire process.
Launching the app was a critical milestone, transitioning LePal from beta testing to reaching a broader audience.
Interview
Interviewed 5 target users
Strategy
Defined the MVP with PM and AI scientist
Design
Created flows and system logic for core features
Usability testing
Created flows and system logic for 3 core features
I translated qualitative patterns into system logic
I collaborated with cross-functional teams to ensure the app met high-quality standards and was ready for deployment.
Also, I prioritized features based on emotional utility and user resonance
System Thinking
Making Empathy Scalable
Why these systems?
Because real emotional connection at scale requires more than good intentions—it demands infrastructure. To make that possible, I:
What We Didn't Build (and Why)
Not every idea makes it into the product—and that’s by design. After early prototyping and cross-functional reviews, we identified two features that, while initially promising, didn't align with our emotional safety standards or the simplicity we were aiming for:
Process
From Insight to Interaction
I translated qualitative patterns into system logic, collaborated across disciplines to prototype interactions, and prioritized features based on emotional utility and user resonance. This was more than just translating insights—it was about designing with emotional sensitivity, under ambiguity, while staying aligned with technical and ethical constraints.
Discovery (Let’s call it the emotional excavation phase)
I started by conducting five semi-structured interviews and a three-day diary study to understand how students really felt while interacting with wellness apps.
5 semi-structured interviews with Gen Z grad students
3-day diary study tracking mood and journaling habits
Mapped emotional drop-off patterns—users often disengaged after Day 3 when the app failed to adapt or acknowledge them
Defining Emotional Design Principles
From this, I distilled 3 key design principles:
Principle #1
Let users lead
offer structure but never control
Principle #2
Speak with care
match tone to user energy
Principle #3
Reward consistency
small rituals over deep work
Prioritizing What Matters Most
We prioritized features based on:
Value vs. complexity: We mapped each feature’s expected emotional impact against implementation difficulty, focusing on features with high resonance and manageable build scope.
Frequency of use (daily rituals preferred): Users told us that small, repeatable moments were more sustainable than deep one-time reflections—so we optimized for habitual features.
Potential for emotional connection: Features were evaluated not just by usability, but by their capacity to create trust, presence, or validation during emotional lows.
Results
Key Systems I Designed
My contributions helped LePal achieve measurable success:
🔍 AI Tone Logic Flow
The AI adjusts its tone using real-time inputs: mood score, time of day, and recent journal sentiment. Each combination maps to a specific tone output: grounding, validating, or celebratory.
“It didn't always cheer me up—sometimes it just sat with me. That felt real.” —Participant #5
📓 Adaptive Journaling Flow
The AI adjusts its tone using real-time inputs: mood score, time of day, and recent journal sentiment. Each combination maps to a specific tone output: grounding, validating, or celebratory.
“It didn't always cheer me up—sometimes it just sat with me. That felt real.” —Participant #5
🔮 Crystal Ball Flow
The AI adjusts its tone using real-time inputs: mood score, time of day, and recent journal sentiment. Each combination maps to a specific tone output: grounding, validating, or celebratory.
“It didn't always cheer me up—sometimes it just sat with me. That felt real.” —Participant #5
🌍 Therapy Planet Topic Flow (Figma diagram placeholder)
The AI adjusts its tone using real-time inputs: mood score, time of day, and recent journal sentiment. Each combination maps to a specific tone output: grounding, validating, or celebratory.
“It didn't always cheer me up—sometimes it just sat with me. That felt real.” —Participant #5
System Thinking: Making Empathy Scalable
To help our engineering team implement these features:
Multi-method field investigations
observations at 4 retail stores
Desk research
Interviews
with 4 employees and 3 customers
What We Didn't Build (and Why)
We scoped out but chose not to implement:
Multi-method field investigations
observations at 4 retail stores
Desk research
Interviews
with 4 employees and 3 customers
Outcomes & Impact
Designing for Connection, Clarity, and Care
My contributions helped LePal achieve measurable success:
%
positive sentiment feedback
driven by empathetic design and better integration of the spirit character.
%
positive sentiment feedback
driven by empathetic design and better integration of the spirit character.
%
positive sentiment feedback
driven by empathetic design and better integration of the spirit character.
Reflections and Learnings
Designing for Meaningful Change
LePal taught me that good product design is about designing relationships—especially when the product is meant to support emotional well-being.
I learned how to translate emotions into structured logic, balance scale with care in tone systems, and prioritize rituals over features.