Emergencies overload people. In a real cardiac arrest or similar crisis, bystanders are often scared, unsure what to do, and forget training. Clear, immediate, step-by-step guidance can bridge the gap between “I once took a class” and action now.
Phones are already there. People reach for their camera and voice first. A coach that uses live video + speech can adapt to what’s actually on screen (“move the camera,” “hands higher,” “keep pushing”) instead of only playing a generic audio script.
Consistency and safety-by-design. Letting a raw chatbot improvise medical steps is risky. A bounded workflow (the app owns the steps; the model mainly interprets and phrases) keeps instructions short, ordered, and on-protocol, which fits how emergency help is supposed to be delivered: one decision at a time.
Augmentation, not replacement. The point isn’t to replace 911 or professional responders—it’s to support the first minutes when every second counts and a calm voice in the ear can change whether someone starts and sustains CPR.
In one line: it exists to turn panic and uncertainty into structured, timely action when a human is alone with a crisis and a phone
Log in or sign up for Devpost to join the conversation.