AITestnet / beta

U-HANDSHAKE

Team Siri Khanansuk

The big news

Gen Z can stop overspending and dodge scams before they happen. U-HANDSHAKE splits a spending wallet into three clear zones: Must Keep for upcoming bills and debts, a Protected Buffer for emergencies, and Safe-to-Spend for everything else. Before a payment, it uses deterministic calculations and explainable machine learning to show the real impact, especially under urgency, social pressure or scam tactics. Users can talk it through with a simulated U-Bot assistant. It's an integration-ready safety layer for wallets and payment apps.

Watch the demo

What it does

A financial decision-safety layer that partitions wallet funds, scores payment risk with deterministic rules and explainable scikit-learn models, and offers a simulated U-Bot voice assistant. Built with Next.js, React, TypeScript, Node.js, Python and Zod, using OpenRouter and ElevenLabs, tested with Vitest and Playwright on Vercel.

Try it yourself

Open live demo View the code See the pitch deck

U-tags

Tap a tag to find similar projects

Meet the team

SKSiri Khanansuk

This is an independent hackathon project. Some features may be prototypes. Please do your own research before relying on it.