TrustGuard-AI

AI that protects job seekers from scams

Comment

TrustGuard AI 🛡️

Inspiration

Online job scams are increasing rapidly, especially targeting students, fresh graduates, and remote job seekers. Fraudulent job postings often ask for registration fees, training payments, or redirect applicants to fake websites and messaging apps like Telegram or WhatsApp.

Many job seekers struggle to verify whether a job opportunity or company is legitimate. Existing job platforms provide limited tools for detecting scams, leaving users vulnerable to financial loss and identity theft.

We were inspired to build TrustGuard AI to protect job seekers by using artificial intelligence to automatically analyze job postings and verify company legitimacy. Our goal is to create a simple tool that helps users make safer decisions before applying for a a job.

What it does

TrustGuard AI is an AI-powered job scam detection platform that analyzes job postings and evaluates their trustworthiness.

The platform performs multi-layered analysis including:

  • Job description analysis
  • Scam keyword detection
  • Domain reputation verification
  • Company website validation
  • AI semantic analysis of recruitment language

The system generates a Trust Score that indicates whether a job opportunity is safe or potentially fraudulent.

Trust scores are calculated using a weighted model:

$$ TrustScore = (JobScore \times 0.6) + (CompanyScore \times 0.4) $$

This helps users quickly identify suspicious job postings and avoid scams.

Key Capabilities

  • Detecting phishing-style recruitment messages
  • Verifying company websites for legitimacy
  • Checking domain mismatches between job links and official company domains
  • Providing explanations for why a job is considered risky

How we built it

TrustGuard AI is built using a full-stack architecture that integrates modern web technologies with AI-powered analysis.

Frontend

  • Next.js 14
  • React
  • TypeScript
  • Tailwind CSS

The frontend provides a clean interface where users can paste a job description or provide a company website URL for analysis.

Backend

  • Python
  • FastAPI
  • Uvicorn

The backend handles job analysis requests, company verification, and trust score calculations.

AI Layer

We integrated Nvidia NIM Cloud AI (Mixtral 8x7B) to perform semantic analysis of job descriptions. The model evaluates the context and language used in recruitment messages to detect suspicious patterns.

Company Verification

We used Firecrawl API and BeautifulSoup to analyze company websites and validate:

  • SSL / HTTPS security
  • Contact pages
  • Privacy policies
  • Social media presence
  • Content authenticity

All these signals are combined into a final risk assessment score.

Challenges we ran into

One of the main challenges was verifying company legitimacy automatically. Many legitimate companies have minimal websites, while some scam websites mimic real companies very well.

Another challenge was job board scraping restrictions. Platforms like LinkedIn and Indeed block automated scraping, so we designed the system to allow manual job description input.

Integrating AI semantic analysis while maintaining fast response times was also challenging. We optimized the system to balance accuracy and performance.

Finally, designing a trust scoring model that combines multiple signals (AI analysis, domain verification, website checks) required careful tuning to reduce false positives.

Accomplishments that we're proud of

We successfully built a working AI-powered prototype that can detect suspicious job postings and verify company legitimacy in real time.

Key achievements include:

  • Implementing a multi-layer scam detection system
  • Integrating Nvidia Cloud AI for semantic analysis
  • Building a combined trust scoring model
  • Developing a modern, user-friendly web interface
  • Deploying a live working demo

The platform demonstrates how AI can be applied to solve a real-world social problem affecting millions of job seekers.

What we learned

Through this project we learned:

  • How to integrate large language models into real-world applications
  • Techniques for detecting scam patterns in natural language
  • Building scalable APIs using FastAPI
  • Combining AI analysis with rule-based verification
  • Designing systems focused on user safety and trust

We also learned that combining AI intelligence with traditional verification methods significantly improves detection reliability.

What's next for TrustGuard AI

We plan to expand TrustGuard AI with several new features:

  • Browser extension for one-click job verification
  • Integration with job platforms and recruitment websites
  • Domain age and WHOIS verification
  • Detection of fake recruiter profiles
  • Crowdsourced scam reporting
  • Multilingual support to protect global job seekers

Our long-term goal is to build TrustGuard AI into a global safety layer for online job searching, helping millions of people avoid scams and find legitimate opportunities safely.

Built With

Share this project:

Updates