RoleFit AI

Inspiration

Inspiration

The idea came directly from my own experience with job hunting. I ran into serious difficulties and found the whole process exhausting, tedious, and painfully slow. At one point, I was just blindly applying to everything and facing endless rejections without any feedback. I wanted to build something that solves this frustration so people don't have to waste their energy on repetitive, manual applications.

What it does

RoleFit AI displays the job vacancies that best fit the candidate. It breaks down each vacancy in detail and provides an exact percentage score showing how well the user matches the role. It also generates a tailored cover letter draft and resume adaptation advice, highlighting how candidates can reframe their existing experience for missing requirements.

How we built it

We built RoleFit AI using Next.js (App Router), TypeScript, and Tailwind CSS. The application runs extremely fast and protects user privacy by parsing uploaded resumes ephemerally in RAM (using pdf-parse and mammoth) without ever saving data to a database or disk.

My favorite part of the build is the deterministic scoring engine: instead of unpredictable AI guesses, it uses clear mathematics to calculate the 50/35/15 match breakdown:

• 50% Skills Match (required vs. preferred skills)
• 35% Experience & Projects (domain alignment and depth)
• 15% Work Format & Location Fit

Challenges we ran into

Designing the user interface was quite challenging—it took serious effort to simplify the layout and make it intuitive without visual clutter. Dialing in the mathematical scoring engine and balancing the 50/35/15 formula so that every score is fair, explainable, and accurate was also a demanding task.

Accomplishments that we're proud of

I am proud that practically everything in the application works exactly as planned. Successfully finishing the project and seeing the entire end-to-end flow run reliably from start to finish is a great milestone.

What we learned

Building this project helped me gain a deeper understanding of how modern web applications work, component architecture, and handling ephemeral data safely in memory.

What's next for RoleFit AI

In the future, we plan to connect RoleFit AI to live job feeds and integrate real automated application submission via official ATS endpoints (like Greenhouse and Lever) or browser extensions, while maintaining candidate review safeguards.

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for RoleFit AI

Built With

• css
• elevenlabs
• next.js
• node.js
• playwright
• tailwind
• typescript