https://docs.google.com/presentation/d/1kRaQyKPKHA3sn7sRTVwOSdsWiQJeeMlTRAeg5QPmgUQ/edit?usp=sharing (see the hardware slide for specifics)

A wearable armband and iOS app that helps you track UV exposure in real time and receive personalized sun safety recommendations.

Too much sun exposure is the leading preventable cause of skin cancer, but most people don’t realize when they’re at risk. We wanted to create a simple, wearable way to monitor UV exposure and make sun safety more accessible—especially across different skin tones.

Sollis is a wearable + mobile system that:

• Uses a UV sensor to measure sunlight intensity.
• Processes data in a backend to calculate a risk score.
• Leverages Cerebras AI to generate personalized insights (e.g., when to reapply sunscreen or step into the shade).
• Displays results in a Swift-based iOS app with actionable guidance.

Hardware: Arduino board + UV sensor programmed in C++ to capture exposure data. Backend: Python service that processes UV readings, calculates risk, and communicates with the Cerebras API. Mobile App: An iOS app written in Swift to visualize exposure data and deliver recommendations. Database & Cloud: Firebase + GitHub integration for data handling and collaboration.ion.on.

• Compiling issues while integrating Arduino code into the broader system.
• Designing a UV-to-risk formula that was both accurate and meaningful.
• Getting each layer of the stack (hardware → backend → AI → app) to communicate reliably under hackathon time constraints.

• Built a full-stack prototype in under 36 hours.
• Successfully integrated hardware, backend logic, AI inference, and mobile UI into a working system.
• Created a user-friendly experience that turns raw sensor data into clear, actionable insights.

• How to connect hardware, backend, and frontend into a cohesive system.
• Debugging integration issues under tight deadlines.
• Refining formulas for real-world health accuracy.
• How AI can make technical health data feel personal and actionable.

• Sleeker, more comfortable wearable design.
• Better sensor calibration for improved accuracy.
• Expanded inclusivity by tailoring recommendations across a wider range of skin tones.

Arduino, C++, Cerebras, Firebase, GitHub, Python, Swift By: Emma Wong, Adishree Das, Vijeta Garg, & Skyler HallHalll