TROT

TROT

Threat Reduction and Oversight Technology

Inspiration

TROT came from thinking about how easily scammers can isolate vulnerable people, especially seniors, during a phone call. A scammer can create fear or urgency and convince someone to act before they have the chance to talk to a person they trust.

We wanted to create something that brings that trusted person into the situation earlier. Instead of expecting a senior to recognize a scam on their own, TROT connects them with a guardian or caregiver who can see what is happening and step in when needed.

What it does

TROT is a two-sided scam protection system for a senior and their trusted guardian.

The senior has a very simple screen that shows whether they are protected, currently on a call, or dealing with a possible scam. The guardian has a more detailed dashboard where they can see active calls, alerts, transcripts, and previous call activity.

The guardian can also connect directly to a senior's account and, during a real active Twilio call, remotely end the call if they believe the senior is in danger.

For our demo, we also built a simulated scam-call flow that updates both sides in real time. The guardian can watch the call move from ringing to active, see transcript messages appear, and receive a scam alert. At the same time, the senior's screen changes from its normal protected state to a clear warning.

How we built it

We built the frontend using Next.js and connected it to Supabase for authentication, user relationships, call information, alerts, and realtime updates.

We used Twilio for our phone system. Calls to a senior's TROT number can be forwarded to their actual phone, allowing them to answer normally. We also built a guardian control that can terminate a real active Twilio call remotely.

The senior and guardian accounts are paired through the guardian dashboard. Once they are connected, that relationship is saved so the senior does not have to repeatedly enter codes or configure anything.

We deployed the frontend using Vercel, which allowed us to open and test TROT directly on our phones and make the web prototype feel much closer to a mobile app.

Challenges

One of our biggest challenges was figuring out how much responsibility should be placed on the senior. Early ideas involved having them manually report something suspicious, but we realized that if someone already trusts the scammer, they may never think to do that.

That pushed us toward a system where TROT and the guardian can react to what is happening instead.

Phone integration was another challenge. We had to work with real Twilio call routing, different call states, and making sure the guardian could end the correct active call without interfering with other call records.

We also learned that the two users need very different interfaces. The guardian needs details and controls, while the senior needs something calm, simple, and easy to understand.

Accomplishments that we're proud of

We're proud that TROT became more than just a UI concept. We were able to connect real phone calls through Twilio, build separate senior and guardian experiences, pair the two accounts, synchronize updates through Supabase, and give the guardian the ability to remotely end an active call.

We're also proud of the senior interface. Instead of filling it with technical information or risk percentages, we focused on making it clear whether everything is okay and what the user should do if something is wrong.

What we learned

We learned that protecting someone from a scam is not only a cybersecurity problem. It is also an accessibility and communication problem.

The person being protected should not need to understand how the detection system works. They need to know whether they are safe and who they can turn to for help.

We also learned a lot about working with realtime data, authentication, Twilio call handling, and designing one system for two very different users.

What's next for TROT

Our next step would be replacing the simulated scam detection with live call transcription and automatic analysis.

The goal is for TROT to eventually recognize suspicious behavior during a real conversation, alert both the senior and guardian, and give the guardian enough context to decide whether they need to step in.

We would also like to expand TROT into a full mobile experience with stronger notifications, smarter scam detection, and more ways for trusted family members or caregivers to protect the people they care about.