AI Resume Analyzer

Full-stack resume analysis app with authentication, PDF upload, Gemini-powered evaluation, and a live dashboard fed by backend API aggregation.

---

Features

• User signup and login with JWT token generation.
• Secure password hashing using bcrypt with long-password-safe preprocessing.
• Resume upload endpoint (PDF only).
• AI resume analysis via Google Gemini.
• Smart Gemini model fallback (auto-selects a supported model if configured model is unavailable).
• Dashboard metrics sourced from real backend data (/dashboard-data), not hardcoded frontend values.

Tech Stack

• Backend: FastAPI, PyMongo, PyPDF2, Google Generative AI SDK, bcrypt, python-dotenv
• Frontend: React, Vite, Axios, Recharts
• Database: MongoDB (Atlas)

Project Structure

AI-resume-analyzer/
	backend/
		main.py
		auth.py
		database.py
		gemini_ai.py
		routes/
			auth_routes.py
			resume_routes.py
		utils/
			pdf_parser.py
	frontend/
		src/
			App.jsx
			api.js
			pages/
				Login.jsx
				Signup.jsx
				Dashboard.jsx
				UploadResume.jsx

Prerequisites

• Python 3.10+
• Node.js 18+
• npm 9+
• MongoDB Atlas connection string
• Gemini API key

Backend Setup

Run in a terminal from project root:

cd backend
python -m venv .venv
.\.venv\Scripts\activate
pip install fastapi "uvicorn[standard]" pymongo python-dotenv PyPDF2 google-generativeai bcrypt "python-jose[cryptography]" python-multipart

Create or update backend/.env:

GEMINI_API_KEY=your_gemini_api_key
GEMINI_MODEL=gemini-2.5-flash
JWT_SECRET_KEY=replace-with-a-strong-secret

Important:

• MongoDB URL is currently defined in backend/database.py (MONGO_URL).
• Replace it with your own connection string before running.
• For production, move MONGO_URL to environment variables.

Start backend server:

uvicorn main:app --reload

Backend URLs:

• API base: http://127.0.0.1:8000
• Swagger docs: http://127.0.0.1:8000/docs

Frontend Setup

Run in a second terminal from project root:

cd frontend
npm install
npm run dev

Frontend URL:

• Vite dev server: http://localhost:5173 (or next available port like 5174)

The backend CORS config allows localhost loopback ports, so alternate Vite ports still work.

API Endpoints

GET /health

Returns backend health status.

POST /signup

Request body:

{
	"email": "user@example.com",
	"password": "StrongPassword123!"
}

POST /login

Request body:

{
	"email": "user@example.com",
	"password": "StrongPassword123!"
}

Response:

{
	"token": "<jwt_token>"
}

POST /upload-resume

• multipart/form-data
• field name: file
• accepts: .pdf only

Response shape:

{
	"analysis": {
		"score": 75,
		"skills": ["Python", "FastAPI"],
		"missing_skills": ["Docker"],
		"suggestions": ["Add quantified project outcomes"],
		"model_used": "gemini-2.5-flash"
	}
}

GET /dashboard-data

Aggregates real data from stored resume analyses and returns:

• overall_score
• ats_pass_rate
• versions_uploaded
• issues_found
• latest_analysis
• chart arrays: radar_data, keyword_data, trend_data, section_data, activity_data
• action_items

How Dashboard Data Works

• Every upload stores { filename, analysis, created_at } in MongoDB.
• GET /dashboard-data calculates dashboard metrics from recent history.
• Dashboard.jsx fetches /dashboard-data on load.
• After a successful upload, UploadResume.jsx triggers a dashboard refresh so charts and counters update immediately.

Troubleshooting

• AI analysis is unavailable:

  • Check GEMINI_API_KEY in backend/.env.
  • Restart backend after editing .env.

• Gemini model error (404/not found):

  • The app auto-falls back to supported models.
  • If needed, set GEMINI_MODEL=gemini-2.5-flash.

• Only PDF files are supported:

  • Upload a .pdf file only.

• Frontend cannot reach backend:

  • Ensure backend is running on port 8000.
  • Confirm frontend uses VITE_API_BASE_URL or defaults to http://localhost:8000.

• Exit code 1 after uvicorn --reload or npm run dev:

  • If you stopped with Ctrl+C or timeout, this can be expected.

Security Notes

• Never commit real API keys or database credentials.
• Rotate keys immediately if they are exposed.
• Use environment variables for all secrets in production.