Roamie is an AI-powered wearable pendant designed to enhance child safety through real-time facial and voice recognition, GPS tracking, and instant emergency alerts. It blends seamlessly into a child’s daily routine, offering proactive protection against stranger danger, accidents, and other risks — while empowering independence and reassuring parents.
My Role
Tools
Duration
Team
UI/UX Designer
UX Researcher
Figma
Miro
Powerpoint
Unbouce
Google Analytics
Jan 2025- May 2025
2 UX Designers
1 UX Researcher
1 Product Strategist
1 Product Engineer
My Impact
I was involved in the full design process — from conducting parent interviews and desk research to creating user flows, prototypes, and branding. I tested the MVP with parents to refine usability and safety alerts, and designed the final poster to communicate the concept. The result was a validated, user-focused safety platform that empowers families with real-time monitoring and peace of mind.
Context
Nearly 41% of parents in the US are stressed about their children’s physical and mental safety. Existing safety solutions are often reactive, limited in scope, and lack real-time, context-aware features. Our goal was to create a solution that proactively detects danger and prevents harm before it occurs.
Problem
Parents often struggle to monitor their children’s safety and well-being when they are out of sight, making it difficult to detect potential risks, respond quickly to emergencies, or ensure children feel secure and independent. Without a discreet, context-aware, and easy-to-use solution, parents are left with incomplete insights into their child’s day-to-day activities and safety patterns.
Key Deliverables
We developed Roamie, an integrated child-safety platform with two key components:
Child Wearable Device – Lightweight, discreet, and equipped with sensors for location tracking, environmental monitoring, and interaction detection.
Parent Mobile App – A real-time dashboard for alerts, location history, safety zone configuration, and interaction reports, enhanced with AI for context-aware insights.
How Might We Statement
How might we enhance child safety with innovative and intuitive technology, enabling rapid emergency response, independence, and a secure family support network?
Solution Preview
Research
Goals: Our research aimed to: Understand parents’ top safety concerns for their children, both in physical and digital spaces. Learn how parents currently address these concerns, and where existing solutions fall short. Gather qualitative insights from parents to guide feature prioritization for a child safety wearable.
Competitive Analysis
As part of our research methodology, we began by surveying a wide range of existing safety and tracking devices on the market. After an initial scan of over a dozen products, we narrowed our focus to three key competitors that represented distinct approaches to child safety technology: Humane AI Pin, AngelSense GPS Tracker, and Apple AirTag.
We evaluated each competitor based on:
Usability – How well the product facilitated real-time location tracking, emergency activation, and monitoring for parents or guardians.
Features – The range and uniqueness of safety features offered, including GPS tracking, SOS modes, audio/visual recognition, and parental controls.
Form Factor & Design – The device’s size, discretion, durability, and child-friendliness.
User Engagement – Strategies to ensure consistent usage and trust, such as seamless alerts, two-way communication, or AI-driven personalization..
Humane AI Pin
AngelSense GPS Tracker
Apple Airtag
Digital or Physical
Hybrid
Physical
Hybrid
Target Age Group
All ages
Ages 3+
All ages
Customization
Target Age Group
Target Age Group
Target Age Group
Price
High ($699+)
High($229 + monthly subscription)
$29 per Airtag or $99 for 4-pack
Key Features
AI assistant, voice control, laser projection
SOS mode, two-way voice communication, fall detection
Precision finding with UWB for accurate location tracking
Customization
Advanced AI, hands-free use, no screen distraction
Strong safety features, real-time tracking, parental monitoring, SOS emergency mode
Small, discrete, easy to attach to belongings; great integration with Apple devices
Customization
Expensive, not child-focused, requires interaction
Needs frequent charging (24–48 hours), high cost, requires a subscription
For tracking belongings, not kids; no two-way communication
Key Takeaways:
Gap in Child-Specific Solutions – Existing products either target general users (AI Pin, AirTag) or partially address child safety but with significant trade-offs (AngelSense).
Need for Feature Integration – No single device combines real-time tracking, SOS functionality, facial/audio recognition, and parental controls in a child-friendly form factor.
Opportunity for Discreet, Always-On Safety – A lightweight, discreet, AI-powered wearable with long battery life and comprehensive safety features would fill a critical gap in the market.
Survey
Format: Combination of quantitative and qualitative questions, including multiple choice, Likert scales, ranking, and open-ended responses.
Recruitment: Distributed via Reddit forums, Facebook parenting groups, printed flyers in family apartment complexes and graduate housing, and personal networks.
Sample:
Total Responses: 46
Gender Breakdown: 39 female, 7 male respondents
Children’s Ages: Ranged from 11 months to 18 years
Purpose: Identify the most common safety concerns and current prevention strategies used by parents, as well as measure demand for proactive safety solutions.
Key Findings from Survey
Most Common Concerns
2 of 3: Mental Health & Digital Safety
3 of 3: Leaving children with others
Interviews
Participants:
5 parents selected from survey respondents who volunteered for follow-up.
Children’s Ages: 15 months to 5 years old.
Approach:
Semi-structured interviews with 8 open-ended questions (covering daily routines, specific safety incidents, trust factors, technology use, and product gaps).
Conducted over Zoom or phone, recorded with participant consent.
Analysis:
Transcribed responses and organized insights using sticky notes in Miro.
Applied affinity mapping to group recurring themes into categories such as “trust & community,” “safety challenges,” “technology use,” and “product frustrations.”
Trust & Community: 4 of 5 only sought childcare from close family/friends due to safety concerns.
Parental Anxiety vs. Independence: Parents felt tension between wanting children to be independent and ensuring constant safety.
Technology Use: Most already used monitoring products for physical safety; all tightly controlled screen time.
Safety Challenges: Physical accidents (falling) were the most common fear, followed by choking and kidnapping.
Product Frustrations: Existing parental controls (especially in entertainment apps) were too limited.
Feature Feedback: All liked the idea of a wearable safety device; differences arose around form factor and use cases.
Defining the Solution
Two storyboards emerged as top contenders
.jpg)
.jpg)
Over time, this technology trains children to recognize safe and unsafe situations, reducing potential risk.
The wearable discreetly ensures the child’s safety with an SOS button, proactive emergency lock, and real-time alerts to parents and a local safety network.
Final Product
We merged the strongest elements into Roamie:
A discreet pendant with facial/voice recognition, GPS tracking, and proactive emergency alerts, supported by a parent app with real-time video, audio, and location data.
Web Probe & Market Validation
Before investing significant time into developing Roamie’s full product ecosystem, we needed to answer a critical question:
Would parents see enough value in a proactive child safety device to take action?
The web probe allowed us to:
Test market interest early without committing to full development
Identify which messaging — independence and growth (positive desire) vs. danger prevention (negative desire) — resonated most with parents
Gauge willingness to engage with our concept through clicks and sign-ups
Process:
Created a single web probe landing page describing Roamie’s concept and features.
Designed two separate Google Ads campaigns:
1. Positive Desire Campaign – Emphasized independence, confidence, and child growth.
2. Negative Desire Campaign – Focused on danger prevention, emergencies, and fear of harm.
Both campaigns targeted parents of children aged 3–12.
Measured CTR, conversion rate, and keyword engagement over a 5-day period.
.jpg)
.jpg)
.jpg)
.jpg)
.jpg)
Key Takeways
Top-performing ad combinations highlight the popularity of real-time location tracking and facial/voice recognition as key product features.
Ads focusing on positive desires (e.g., “independent child,” “relaxed parents,” “child development”) outperform those highlighting negative scenarios, showing that parents respond better to uplifting and reassuring messaging.
Men aged 25-35 engaged the most with our ad content. Additional insights could be uncovered by interviewing this demographic in future user testing.
Overall, our ad was well-received with 11,719 total impressions and a low average cost-per-click of $0.06.
Branding & Mood Board
From the start, we wanted Roamie to feel trustworthy for parents while being approachable and friendly for kids.
The branding needed to strike a delicate balance — professional enough to be taken seriously in a safety context, yet warm enough to avoid feeling like surveillance.
We explored color palettes that communicated safety and clarity — vibrant accents for important alerts, with a calm and neutral base to avoid overwhelming the interface. Typography was chosen for legibility, with rounded edges to maintain a friendly tone.
Mid-fidelity Prototype
.png)
.png)
Notification – Lock-screen alert showing safety concern, child’s name, and location in real time.
Safety Alert – Live video feed with emergency status, location, and quick actions like reporting or calling.
Dashboard – Real-time child status, location, and safety features with a live map view for quick tracking.
Connect – Map and profiles to connect with trusted local Roamie community members.
Device Details – Individual pendant view showing connection status, battery, family network, and saved locations.
Devices – Overview of connected Roamie pendants, battery status, and quick access to monitoring history.
Value Flow- Diagram
Before jumping into MVP testing, we mapped out the value exchange between all stakeholders — child, parent, and platform.
This diagram highlighted where value was created at each interaction point — real-time alerts for parents, discreet wearable for children, and analytics for improving the system. It became the blueprint for prioritizing features in MVP development.
MVP Testing – Experience & UI
Who
Two parents with children under 10 were shown a video of what they would see with Roamie.
Parent 1: Lives in Singapore, child is 2 years old
Parent 2: Lives in California, child is 8 years old
What
As we did not wish to put children in harm’s way, we enacted out a situation where a stranger approaches a child at a park. This recording was then shown to parents along with our lo-fi prototypes.
Where
The two interviews were conducted virtually on Zoom.
Why
The purpose of having these interviews is to collect feedback and insights on Roamie’s initial offerings. By allowing our earlyvangelists to see what they can achieve with Roamie, we gained meaningful input on how they interact with our product, allowing us to iterate.
Process
We enacted out two situations which were identified as top concerns for parents during our primary research:
Stranger Danger: A child playing in a park is approached by a stranger to go with them to a “better” more enticing park. As soon as the stranger says the phrase “ come with me”, the audio recognitions catches it as a red fla and sends an alert to the parent.
Physical Injuries/Accidents: A child playing in the park trips and falls and Roamie senses the fall as the camera falls and alerts the parents. It senses a increase in heart rate and shakiness of the video feedback.
This footage was then shown to parents along with our lo-fis. The parents were asked to imagine their kids in the video and react to the alert notification as well as the tracking screen in accordance with the situation.
Questions asked:
How was your experience with Roamie?
What did you like and dislike about the interaction?
What feedback or concerns do you have about the product?
Key Findings
From the interviews, we uncovered four core themes:
Safety & Privacy Concerns – Parents wanted full control over camera activation, no unnecessary data storage, and separate alerts for physical injury vs. stranger danger.
Pendant Design & Comfort – Needed to be lightweight, waterproof, hypoallergenic, with multiple wearing options and a locking mechanism.
Usability & Accessibility – Minimal setup, auto-pairing, one-tap emergency location sharing, and a clean dashboard.
Pricing & Maintenance – Affordable ($50–$100), multi-pack discounts, and low-cost replacements.
Design Implications:
Balance Safety with Privacy: Real-time monitoring with parental control.
UI Updates: Tiered alert system, essential alerts only, shareable location with trusted adults.
Affordability: Pricing competitive with tracking alternatives.
Seamless Experience: Effortless setup and daily use.
Adaptability: Multiple wearable formats without losing durability.
Final Solution
Proactive Safety Alerts (Safe v. Dangerous)
Personalized Trusted Network to Support Your Child’s Safety
Customizable, Real-time Location Tracking & Sharing
Daily Activity Recaps & Teachable Moments
.jpg)
Final Poster
.png)
Reflection
Funded by the Catalyst Funding Award, this project was a turning point in my growth as a designer. It challenged me to navigate the entire product cycle — from research and ideation to prototyping, MVP testing, and final delivery — while making decisions that balanced user safety with usability. I learned to translate real user concerns into actionable design changes, work iteratively based on feedback, and communicate the product vision through both visuals and storytelling. This experience deepened my confidence in tackling complex, end-to-end design challenges.
RELATED WORK
.png)
.jpg)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)