JobMatch AI

Finding real remote jobs is hard. I built JobMatch AI to help job seekers quickly check if a job is a good fit or a possible scam using AI-powered match scoring, skill analysis, and interview tips.

Comment 1

Inspiration

Remote job boards are flooded with low-quality and fraudulent postings. Job seekers especially newcomers to Canada waste hours applying to fake roles or sharing personal information with scammers. There was no fast, free, AI-powered tool to verify job quality before applying. JobMatch AI solves this by combining resume analysis, scam detection, and career coaching into one simple tool that works in seconds.

What it does

Core Features • Match Score 0-100: AI compares your resume to the job description • Scam Detection: Flags red flags like Telegram-only contact, check deposits, instant hire • Skill Gap Analysis: Shows exactly which skills you are missing • Interview Tips: 3 personalized tips based on the specific job • Resume Suggestions: ATS-friendly improvements tailored to the role • Results in under 10 seconds using Gemini 2.5 Flash

How we built it

I built JobMatch AI solo over the hackathon weekend starting from a Google AI Studio template and customizing it into a fully functional product. Here is the build process step by step: • Started with Google AI Studio to scaffold the React + TypeScript frontend and connect Gemini AI •Engineered the AI prompt carefully to return structured JSON covering match score, scam signals, skill gaps, interview tips, and resume suggestions • Built the frontend UI using React, TypeScript, and Tailwind CSS with animated result cards and color-coded skill tags •Solved Gemini API quota and model version issues, migrating from gemini-2.0-flash to gemini-2.5-flash for new account compatibility • Deployed the production build to Firebase Hosting using the Firebase CLI — fully live at mf-job-match-app.web.app

Challenges we ran into

Challenges i Faced • Gemini API quota limits on free tier solved by upgrading to a paid account • Model version compatibility gemini 2.0-flash deprecated for new accounts, migrated to gemini-2.5-flash • Deploying a React + Vite app to Firebase Hosting for the first time with limited coding experience • Structuring AI prompts to return reliable JSON output for scam detection and skill analysis • Completing a full-stack AI deployment within a tight hackathon deadline

Accomplishments that we're proud of

Building a complete AI powered web application solo within a limited hackathon timeline using React, Firebase, Tailwind CSS, and Gemini AI. I’m also proud that the project focuses on solving a real world problem by helping job seekers identify safer remote opportunities and avoid suspicious job postings.

What I learned

How to integrate Google Gemini AI into a React app using structured prompt engineering and JSON output parsing • How to build and deploy a React + Vite + TypeScript app to Firebase Hosting using the Firebase CLI from scratch •How API quotas, billing, and model versioning work in Google Cloud — and how to debug and fix them under deadline pressure •How to scope a hackathon project — what to build, what to skip, and how to prioritize a working demo over perfect code •That AI tools like Claude can dramatically accelerate solo development — from architecture decisions to debugging to deployment steps

What's next for JobMatch AI

Future Roadmap • LinkedIn job post integration — analyze jobs directly from LinkedIn • PDF resume upload and parsing • AI Career Coach — ongoing guidance based on your profile • Company reputation scoring

Built With

  • ai-assisted
  • built-with-react
  • development
  • firebase
  • firebase-storage
  • firestore
  • for
  • framer-motion-animations
  • gemini-ai
  • github
  • google-ai-studio
  • mvp
  • prototyping
  • rapid
  • tailwind-css
  • typescript
  • workflows
Share this project:

Updates

Submission history