• 

  Logo_image

Guardrail

Inspiration

Searching for a job has become increasingly difficult, and job seekers are being targeted by sophisticated scams that imitate legitimate employers. Fake recruiters, phishing attempts, and fraudulent job postings can lead to financial loss, identity theft, and wasted time. I wanted to build a tool that helps people evaluate opportunities before they invest their time or share sensitive information.

What it does

Guardrail is an AI-powered employment risk assessment tool that analyzes job postings and highlights potential warning signs. Rather than simply labeling a posting as "safe" or "unsafe," it explains the factors that contributed to its assessment so users can make more informed decisions.

The goal is to give job seekers an extra layer of confidence by identifying common scam patterns, suspicious language, missing information, and other indicators that may deserve closer attention.

How I built it

I built Guardrail as a web application that combines a simple user interface with an AI model to analyze job postings and generate structured risk assessments. The application processes the text of a job listing, evaluates potential risk indicators, and returns an explanation designed to help users understand the results.

The project focuses on combining AI reasoning with a practical, real-world problem that affects thousands of job seekers every day.

Challenges I ran into

One of the biggest challenges was balancing helpfulness with accuracy. Not every unusual job posting is fraudulent, so the system needs to communicate uncertainty instead of making absolute claims. Another challenge was designing prompts that produce consistent, explainable results while avoiding unnecessary false positives.

What I learned

Building Guardrail reinforced the importance of explainability when using AI for decision support. Users are more likely to trust AI when it clearly explains why it reached a conclusion instead of only providing a score or recommendation.

I also learned more about prompt design, structuring AI output, and building applications that solve practical problems rather than simply demonstrating AI capabilities.

What's next for Guardrail

Future improvements include:

• Resume-to-job compatibility analysis
• Recruiter credibility assessment
• Company reputation and risk analysis
• Interview safety guidance
• Personalized recommendations for safer job searching
• Support for additional employment marketplaces and recruiting platforms

What's next for Guardrail

Guardrail is intended to evolve beyond scam detection into a comprehensive AI-powered employment risk platform. Future enhancements include:

Resume-to-job matching to identify gaps between a candidate's experience and job requirements. Recruiter credibility analysis using publicly available signals to help identify suspicious outreach. Company risk profiles that summarize information such as employee reviews, hiring patterns, and publicly reported concerns. Interview preparation with AI-generated questions, suggested responses, and company-specific insights. Privacy and identity protection by warning users before they share sensitive personal information. Personalized risk scoring that adapts recommendations based on a user's experience level and career goals. Support for internships, freelance work, and contract opportunities, where scams are increasingly common. Expanded AI explainability, providing clear reasoning behind every recommendation so users understand the factors influencing the assessment.

The long-term vision is to make Guardrail a trusted AI companion that helps people navigate every stage of the job search with greater confidence and safety.

Built With

• app
• css
• next.js
• openai
• router
• tailwind
• typescript