CESYDE

Inspiration

We’ve seen firsthand how stressful hospitals can be, especially when patients are scared or in pain and can’t easily get the attention they need. Nurses are doing their absolute best, but they’re constantly juggling so many things at once. We built Beacon by CESYDE (Sea-side) because we wanted to make those moments a little easier , to give patients a simple way to ask for help, and to give nurses a clearer picture of what needs attention.

What it does

Beacon lets patients press a physical button to request things like water, food, bathroom help, a blanket, or to signal an emergency, without having to wait or struggle to get someone’s attention. The nurse has a small device that also flashes a specific colour for a specific request. On the nurse’s side, Beacon sends all those requests to an AI-supported dashboard that sorts, prioritizes, and gives quick insights so they can respond faster and stay organized.

How we built it

We connected two Arduinos, one for reading button presses and one for feedback, then built a Python + Flask backend to handle all the communication. On top of that, we added an AI layer using Gemini to help organize and rank the requests.

Challenges we ran into

We fought with API model limits, serial port permission errors, hardware not talking to software, inconsistent JSON from the AI, and lots of tiny timing issues. Pretty much every part broke at least once, but we got it working in the end.

Accomplishments that we're proud of

We’re really proud that we built something physical that actually works and could help real people. Seeing the hardware, the backend, and the AI all functioning together.

What we learned

We learned how to make hardware and software work together without losing our minds, how to design AI prompts that stay predictable, and how important simplicity is. Most of all, we learned how much impact a small idea can have.

What's next for CESYDE

We want to make Beacon wireless, support multiple patients at once, add mobile alerts for nurses, integrate it with existing hospital systems, and explore accessibility options like voice-based requests. We see a lot of potential, and we’d love to keep building on it.