Inspiration
Every year, thousands of students lose money to fake internship scams operating on Telegram and WhatsApp. We realized that existing tools rely on black-box AI, which is too slow and unpredictable for a security check.
What it does
InternShield is a browser-based tool where users paste a job URL to get an instant, 6-step security audit. It calculates a risk score by scanning the live webpage for illegal deposit requests, urgency tactics, and unverified contact info.
How we built it
We built a highly optimized, 100% client-side app using React, Vite, and Tailwind CSS. The UI features a custom WebGL moving gradient and LiquidGlass refraction, paired with a deterministic regex rule engine (R-01 to R-15) that analyzes live text fetched via a CORS proxy.
Challenges we ran into
Browser CORS policies strictly block client-side fetching of arbitrary websites, making live scans difficult. We had to build a robust proxy fallback system and a local "Honeypot" environment to guarantee the rule engine could be tested live on stage.
Accomplishments that we're proud of
We are incredibly proud of achieving a "Zero-AI, 100% Explainable" architecture. Every red flag shows the exact regex rule and the matched evidence string, giving users absolute trust in the verdict without relying on LLM hallucinations.
What we learned
We learned how to balance heavy WebGL visual effects with strict performance optimization so the app never lags. We also gained deep hands-on experience with regex-driven security heuristics and handling real-world CORS limitations.
What's next for InternShield
The immediate next step is partnering with InternWell to integrate our rule engine directly into their job board. This will allow us to automatically pre-screen and block malicious listings before a student ever clicks on them.
Log in or sign up for Devpost to join the conversation.