Aeyesafe inc | B2B | SaaS
Aeyesafe is an AI-powered smart monitoring system that helps care homes keep seniors safe through non-wearable technology.
Built for B2B environments, the platform uses a network of sensors (thermal, radar, LiDAR, and sleep sensors) to detect abnormal behavior, monitor well-being, and alert caregivers in real-time. Unlike traditional systems, Aeyesafe doesn’t rely on wearables or cameras - making it ideal for sensitive care settings.
Created 0–1 Alert System for Safer Senior Care
PROBLEM:
Due to poor alert prioritization and volume overload, administrators often miss time-sensitive alerts, putting resident safety and care response at risk.
BUSSINESS GOAL
Deliver finalized, developer-ready designs by mid-August to support a Beta release, ensuring a successful 1.0 launch by mid-October 2023.
PROJECT TYPE
0-1
OUTCOME
By mid-August, the final designs were delivered to developers with clear documentation, ensuring a smooth transition into development.
MY ROLE
Product Designer (UX, Research, UI)
TEAM
Two other Product Designers, Project Manager, Two Engineers
TIMELINE
July 2023 - August 2023
THE CHALLENGE
Building a robust Alert System within a tight 6-week timeline required fast, focused decision-making. I prioritized rapid ideation and took a phased approach - delivering early, then iterating quickly based on feedback.
I collaborated closely with developers and the PM from day one to define key micro-interactions, account for edge cases, and ensure seamless data validation. This alignment helped to balance speed with thoughtful UX, enabling a successful launch under pressure.
To maintain focus and efficiency, I created a structured roadmap guiding the design process.
Who are Aeyesafe users?
Meet James, a care home administrator overwhelmed by a growing volume of incoming system alerts.
With multiple residents to oversee and limited staff resources, it’s becoming harder for him to identify which alerts are urgent and require immediate action. He’s looking for a software solution that can help him prioritize alerts, reduce noise, and ensure nothing critical gets missed - without adding to his administrative burden.
RESEARCH
Since there wasn’t much existing insight into the challenges care homes faced, I took the lead on user research to better understand their workflows and needs. I wanted to make sure I was designing based on real-world problems, not assumptions.
I wrote open-ended interview scripts to guide conversations and supplemented them with desk research from articles and industry reports. This helped uncover what was slowing people down and what they actually expected from the system.
The insights shaped the product direction early on and gave the team a clearer picture of where to focus our efforts.
Methods
Secondary research and semi-structured user interviews
Research Findings from Secondary Research
50%
Exceeded expected response times
Participants
5 care home administrators
Timeline
1 week
Research Findings from User Interviews
The real challenge for care homes isn’t just the number of alerts - it’s the risk of missing critical ones and the limited time they have to respond.
With so many notifications coming in, staff can feel overwhelmed or unsure which ones matter most. This puts residents at risk if critical alerts are buried under routine updates or not clearly prioritized. And with limited time and staff capacity, even a few seconds of delay can make a big difference in urgent situations.
“When an alert comes in without context, I have to stop what I’m doing to track down the details. That slows everything down.”
10%
of calls canceled
over 3%
of calls forgotten entirely
Lack of context delays action - while new alerts keep piling up.
When alerts don’t clearly explain what’s happening or what to do next, staff waste time digging for details or asking around for clarification. This slows down response times, adds frustration, and makes it harder to keep up as more alerts continue to come in.
“It’s overwhelming - by the time I understand one alert, there’s already a backlog of others waiting.”
“If I had the right information upfront, I could prioritize and act faster instead of playing catch-up.”
How might we help care home administrators quickly identify and respond to urgent alerts among a flood of incoming system notifications?
GUIDING PRINCIPLES
🔺 Prioritization – Design alerts for quick scanning, making critical issues stand out clearly while minimizing cognitive load.
⚡ Actionability – Alerts should be easy to understand and act on, enabling quick responses and eliminating unnecessary steps.
🧩 Context – Each alert should present key details upfront, giving users the clarity they need without requiring additional searches.
IDEATION
To translate the research findings into actionable insights, I created two key artifacts:
Storyboard
High-level user flow.
Storyboard: The Resident-to-Resolution Journey
The storyboard visually mapped the sequence of events from the moment a resident experiences a critical issue to the point of resolution by the care home administrator.
This helped align the team on the real-world context behind the alerts and surface critical moments in the workflow.
User Flow: System-Level View
In parallel, I developed a high-level user flow to understand how users would interact with the system internally. This flow mapped entry points, branching logic based on alert types, and quick-action paths. It helped pinpoint two core decision moments:
How multiple alerts are grouped or collapsed
How fast assignment and follow-up actions are triggered from each alert card
These flows served as a blueprint for our information architecture and interaction patterns.
DESIGN EXPLORATIONS + DISCOVERY WORKSHOP
Turning insights into low-fidelity prototypes, I prioritized the alert notifications first - users’ biggest pain point. Instead of overwhelming them with vague alerts, I aimed to surface the right information at the right time.
I partnered with two product designers to explore early design directions and ran co-creation workshops with three administrators to validate assumptions and find the right balance between clarity and actionability.
DESIGN
Once alert notifications were finalized, I moved on to the Alert page - the central hub for monitoring and action. Using insights from user interviews, I collaborated with other product designers to refine the layout, highlight critical information, and streamline quick actions.
I also worked closely with engineers to ensure the page was scalable, intuitive, and easy to navigate, enabling administrators to manage alerts with confidence.
The next step was to ensure a cohesive, end-to-end experience by addressing dependencies across the system. I began designing the intermediate steps and micro-interactions that connect key moments in the alert workflow - focusing on transitions, confirmations, and user feedback.
In parallel, I worked through edge cases and crafted clear, actionable error messages to support users in high-pressure scenarios. This attention to detail helped reduce confusion, build trust in the system, and ensure the experience remained smooth and reliable, even when things didn’t go as planned.
Alert Filters
Administrators were overwhelmed by too many alerts and no clear way to prioritize. We designed flexible filters by type, time, resident, and urgency - so they could quickly find what matters and act faster.
Resident Alerts
ALERT SYSTEM
We made it easier to monitor each resident by showing alert history and status directly on their profile, giving administrators quick, context-rich insights without switching screens.
Alert Reports
To help administrators spot patterns and improve care, we added reports that surface trends across the past 12 hours, week, or month, broken down by alert type or resident.
Custom Thresholds
Since every care home has its own protocols, we made alert thresholds adjustable, so teams can set the right level of sensitivity for their needs.
Key Learnings
Better Prioritization Improves Usability – Categorized alerts and contextual info reduce response time.
Privacy vs. Safety Balance – AI-driven monitoring must protect privacy while ensuring safety.
Although I left before the official launch, Beta Version 1.0 of the alert system was built on the framework I designed, ensuring long-term impact and future scalability.
Next Steps
AI Enhancements – Conduct further AI model training to improve alert accuracy and reduce unnecessary notifications. Implement machine learning to recognize false positives and refine alert categorization.
Broader Research – Gather insights from caregivers and diverse care homes.
Impact Measurement – Track response times and user satisfaction.