Fit Check AI

Your resume says you're qualified. FitCheck AI tells you if you're actually the right fit instantly.

Comment

🎥 Watch the demo video

Inspiration

I'm currently job hunting, and one of the biggest frustrations I kept running into was never knowing if I was actually a fit for a role before applying beyond just matching keywords on a resume. I'd spend hours on applications only to wonder afterward whether I was even the right fit in the first place. So I decided to build a tool I could use myself: something that gives an honest, data-backed read on fit before you invest hours into an application.

What it does

FitCheck AI analyzes a user's background against a target role or career path and returns a clear, actionable fit score highlighting strengths, gaps, and specific next steps to close those gaps. Instead of a vague "you're a good match," it breaks down why, so users can make informed decisions about which roles or industries to pursue.

How we built it

I built FitCheck AI using Lovable for rapid AI-assisted frontend development, paired with Supabase for authentication, database, and backend logic. The app was deployed on Vercel, with Google OAuth integrated for seamless sign-in. This stack let me move fast from concept to a working product within the hackathon timeline while keeping the architecture clean enough to extend post-submission.

Challenges we ran into

One of the biggest hurdles was getting Google OAuth to work correctly on my Vercel deployment. Lovable's auth wrapper generates a non-standard callback path that isn't natively compatible with external hosting environments, which meant debugging the redirect flow at a level below what the AI tooling exposed by default. I also had to think carefully about how to migrate off Lovable Cloud toward a self-owned Supabase project without disrupting functionality a step I'm tackling post-submission.

Accomplishments that we're proud of

Getting a full AI-assisted analysis pipeline from user input to a structured, meaningful fit report working end-to-end within the hackathon window, solo. I'm also proud of shipping a genuinely useful tool, not just a proof of concept, with real authentication and a production deployment.

What I learned

Working across Lovable and Supabase taught me a lot about the tradeoffs of AI-assisted development tools: they accelerate the easy 80%, but the last 20% like non-standard auth flows still requires manual debugging and a solid understanding of what's happening under the hood. But beyond the technical lessons, this project reinforced something bigger for me: when you build based on a problem you've actually faced, you're not just building a product you're building a meaningful solution. That's the whole point of building. It's not about shipping features for the sake of it; it's about making an impact and making someone's life a little easier, starting with your own.

What's next for Fit Check AI

Migrating fully off Lovable Cloud to a self-owned Supabase project for more control and scalability, expanding the analysis model to cover more career paths and industries, and adding features like personalized upskilling roadmaps based on identified gaps.

Built With

Share this project:

Updates

Submission history