DEV Community

Cover image for I Built a Job Scam Detector in 48 Hours Using Only Conversation
Barbra Atieno
Barbra Atieno

Posted on

I Built a Job Scam Detector in 48 Hours Using Only Conversation

#nocode #ai #hackathon #builtwithmedo

The backstory

A friend lost $150 to a fake job offer last month.
Professional LinkedIn. Convincing offer letter. Gmail recruiter address.

All fake.

I built TrustHire AI so that never happens again to anyone.

What is TrustHire AI?

A full-stack web app that helps job seekers verify whether a
job offer is legitimate or a scam before they share personal
data, pay fees, or accept offers.

Core features:

  • πŸ”΄ AI Risk Scoring Engine (0–100% fraud score)
  • 🚨 Signal Detection (personal emails, payment requests, salary anomalies)
  • πŸ“‹ Community Scam Reports database
  • πŸ“Š User dashboard with analytics
  • πŸ’Ž Premium glassmorphism UI

The Tech Stack

Everything: MeDo.

No React. No Node. No database setup. No deployment config.
Just structured natural language prompts and a working app.

How the Risk Engine Works

I defined scoring rules in plain English. MeDo built the logic.

Signal Detected Risk Score Added
Gmail/Yahoo recruiter email +25 points
Upfront payment requested +35 points
Unrealistic salary +20 points
No company website +20 points

Score 0–30 = βœ… Safe

Score 31–60 = ⚠️ Suspicious

Score 61–100 = 🚨 High Risk

Live Demo

πŸ”— https://app-bphw5ze0v18h.appmedo.com

#BuiltWithMedo

Top comments (0)