RecruiterCheck

Inspiration

We were inspired by the floods of messages from fake recruiters that were appealing to us as students looking for internships that we were about to fall for.

What it does

Uses a fine-tuned LLM to search for patterns and analyze and determine the percent chance of a message being fake or real.

How we built it

We used google collab to fine the LLM and coded on antigravity IDE. Used Next.JS for frontend and FastAPI for backend.

Challenges we ran into

Visual problems, UI not working properly, model not being trained enough to give accurate responses.

Accomplishments that we're proud of

Accuracy of model and feedback and a useful impact.

What we learned

Fine-tuning, web design, creating solutions to real world problems.

What's next for RecruiterCheck

We might train our own model rather than fine tuning to make the model even more accurate and make RecruiterCheck a plug-in for people to use in the future.

Built With

• antigravity
• colab
• fastapi
• groq
• html
• javascript
• next.js
• tailwind