Inspiration
SoundShield didn’t start as a hackathon prompt for us. It started as personal heartbreak. One of our team members lost every single dollar they had to a scam. We know the exact panic, the manipulation, and the gut-wrenching silence that follows. We built SoundShield because once you know what that feels like, you do everything in your power to make sure no one else ever does.
What it does
SoundShield learns what normal looks like for each individual member: their usual payment amount, time, and methods, and steps in before money leaves the account.
Routine payments pass through instantly, but the moment something looks unusual, like a sudden high amount of money transfer or payment to a new recipient. SoundShield triggers a smart pause. It asks the member 3 to 5 targeted questions. The suspicious message will be sent to AI to analyze the possible scam on the spot and requires verified OTP confirmation.
How we built it
Backend: Python with FAST API
Frontend: React with Vite
Challenges we ran into
We couldn't decide which feature was the most important when we came up with more than just a message checker. After talking with the Sound Credit Union staff, we got more ideas on how to detect the scam better based on personal behaviors with AI analysis.
Accomplishments that we're proud of
Setting the time and amount restrictions based on personal behaviors. Also, a 24/7 AI Chatbot and panic button.
What we learned
Many features that we came up with, some of them already exists. We asked the professional to know what we should focus on in a short period of time
What's next for SoundShield
We want to implement it in multiple languages that might be more useful to people who do not speak English as their native language to understand and deal with the problems better.
Log in or sign up for Devpost to join the conversation.