GatekeepeRShield

An AI-powered job scam detector that analyzes job posts and identifies suspicious indicators, providing a risk score and clear reasons to help users avoid fraudulent opportunities.

Comment

Inspiration

Online job opportunities have become an important way for students and job seekers to find employment. However, fake job advertisements can look convincing and may ask applicants for registration fees, processing charges, deposits, or personal information.

We were inspired to build GatekeepeRShield to provide users with an additional layer of protection before they interact with a suspicious job advertisement.

What it does

GatekeepeRShield is an AI-powered job scam detection system. Users can provide job-post text or a screenshot of a job advertisement. The system analyzes the content for suspicious patterns and trust indicators and produces a risk score, classification, and understandable reasons behind the result.

The system categorizes opportunities into:

  • 🟢 Likely Safe
  • 🟡 Suspicious – Verify Before Proceeding
  • 🔴 High Scam Risk

Instead of simply saying that a job is safe or unsafe, GatekeepeRShield explains the indicators that influenced its result, helping users make a more informed decision.

How we built it

The project combines AI-based text analysis with rule-based scam indicators and verification-oriented trust signals. Important indicators include requests for registration or advance payments, urgent hiring claims, personal contact channels, unrealistic income promises, and missing or unverifiable company information.

The system processes the submitted job information and generates a structured result containing the risk level, risk score, and reasons.

Challenges

One of the biggest challenges was distinguishing suspicious job advertisements from legitimate opportunities because scam posts can use professional-looking language and branding.

We addressed this by considering multiple indicators rather than depending on a single keyword. We also designed the output to explain the reasons behind the classification so that users can review the warning signs themselves.

What we learned

Through this project, we learned how AI can be applied to a practical safety problem and how combining automated analysis with explainable results can make an AI system more useful to users.

Future improvements

We plan to improve the system with stronger company and recruiter verification, a larger and continuously updated scam-pattern database, multilingual support, improved image/OCR analysis, and more robust evaluation against real-world job advertisements.

Built With

Share this project:

Updates