Veldr.io

Veldr.online

🌟 Inspiration

Phone scams targeting seniors are heartbreaking and sadly on the rise. I was inspired by seeing elderly loved ones struggle to identify fraudulent calls and the devastating consequences these scams can have on their savings and independence. I wanted to build something proactive that could intervene in real time — before it was too late.

🚀 What it does

Veldr.io is an AI-powered voice assistant that detects suspicious phone conversations in real time, speaks naturally to verify the situation with the senior, and instantly alerts caregivers or trusted contacts. It acts as a digital safeguard, helping protect elders from scams and giving caregivers peace of mind.

🛠️ How we built it

Tech stack: Node.js, Express, MongoDB, React, TailwindCSS
Voice AI: ElevenLabs TTS with SSML for natural, empathetic voice confirmations
Alerting: Twilio for instant caregiver SMS/email notifications
Automation: n8n for routing events and managing escalation workflows
Hosting: Vercel for the frontend, Railway for the backend

The MVP includes a caregiver dashboard, elder profile management, a voice log of flagged calls, and a real-time alert system with escalation.

⚡ Challenges we ran into

Making the voice agent sound human and emotionally supportive rather than robotic
Balancing a fast build timeline with a maintainable, modular codebase
Designing an intuitive UI for caregivers with varying levels of tech literacy
Creating transparent, explainable AI logic that caregivers can trust

🏆 Accomplishments that we're proud of

Prototyping a working MVP in record time
Building a transparent, explainable AI workflow that caregivers can understand
Integrating a real-time alert system to actually prevent scams instead of reacting afterward
Receiving positive early feedback from potential users

📚 What we learned

Caregivers truly value even simple proactive protections
Voice AI can be powerful if designed carefully with empathy and clarity
Hackathons are a fantastic way to validate critical features and assumptions at near-zero cost
Working in public builds momentum and accountability