Professional-grade resume analysis platform powered by advanced AI models with intelligent document type detection and comprehensive insight generation.

Live Demo API Status

An enterprise-grade document analysis platform specializing in resume processing with AI-powered insights. The system automatically detects document types and provides structured analysis optimized for professional recruitment and career development workflows.

• Resume Analysis: Specialized parsing for professional documents with structured output
• Document Type Detection: Intelligent classification of uploaded documents
• Multi-AI Integration: Dual AI provider setup with intelligent fallback mechanisms
• Real-time Processing: Asynchronous document processing with live progress tracking

graph TB
 subgraph "Frontend Layer (Vercel)"
 A[React 18 + TypeScript]
 A1[TailwindCSS Styling]
 A2[Clerk Authentication]
 A3[Mobile Detection]
 A4[404 Error Handling]
 end
 
 subgraph "API Gateway (Vercel Serverless)"
 B[FastAPI Backend]
 B1[JWT Validation]
 B2[File Upload Handler]
 B3[Error Management]
 B4[Health Monitoring]
 end
 
 subgraph "Document Processing Engine"
 C[PDFPlumber Extractor]
 C1[Document Type Detector]
 C2[Content Sanitization]
 C3[Text Preprocessing]
 end
 
 subgraph "AI Analysis Layer"
 D[Sarvam AI Primary]
 E[Gemini AI Secondary]
 F[Intelligent Fallback]
 G[Response Structuring]
 end
 
 subgraph "Data Layer"
 H[MongoDB Atlas]
 H1[User Document Store]
 H2[Binary PDF Storage]
 H3[Analysis History]
 end
 
 subgraph "Authentication & Security"
 I[Clerk Identity Provider]
 I1[JWT Token Management]
 I2[User Session Handling]
 end
 
 A -->|HTTPS API Calls| B
 B -->|Document Upload| C
 C -->|Extracted Text| D
 D -->|Fallback on Error| E
 E -->|Final Fallback| F
 D -->|Success Response| G
 E -->|Success Response| G
 F -->|Keyword Analysis| G
 G -->|Structured Data| H
 B -->|Auth Validation| I
 H -->|User Data| B
 B -->|JSON Response| A

• Node.js 18+ and npm
• Python 3.9+
• Git for version control
• MongoDB Atlas account (free tier available)
• Clerk account for authentication
• AI API Keys (optional - fallback available)

git clone https://github.com/TechyCSR/AI-Powered-Document-Insight-Tool.git
cd AI-Powered-Document-Insight-Tool

cd backend
# Create virtual environment
python -m venv venv
# Activate virtual environment
# On Windows:
venv\Scripts\activate
# On macOS/Linux:
source venv/bin/activate
# Install dependencies
pip install -r requirements.txt
# Setup environment variables
cp env.example .env

Edit backend/.env with your configuration:

# MongoDB Configuration
MONGODB_URI=mongodb+srv://username:password@cluster.mongodb.net/document_insights?retryWrites=true&w=majority
# Clerk Configuration
CLERK_SECRET_KEY=sk_test_your_clerk_secret_key_here
# AI API Keys (Optional - fallback available)
SARVAM_API_KEY=your_sarvam_api_key_here
GEMINI_API_KEY=your_gemini_api_key_here
# Application Settings
ENVIRONMENT=development
DEBUG=True
ALLOWED_ORIGINS=http://localhost:5173,http://localhost:3000

cd frontend
# Install dependencies
npm install
# Setup environment variables
cp env.example .env.local

Edit frontend/.env.local with your configuration:

VITE_CLERK_PUBLISHABLE_KEY=pk_test_your_clerk_publishable_key_here
VITE_API_URL=http://localhost:8000

Terminal 1 - Backend API:

cd backend
# Ensure virtual environment is activated
uvicorn app.main:app --reload --host 0.0.0.0 --port 8000

Terminal 2 - Frontend Application:

cd frontend
npm run dev

• Frontend: http://localhost:5173
• Backend API: http://localhost:8000
• API Documentation: http://localhost:8000/docs
• Health Check: http://localhost:8000/api/v1/health

1. Backend Changes: Auto-reload enabled with --reload flag
2. Frontend Changes: Hot Module Replacement (HMR) active
3. Database: MongoDB Atlas automatically syncs
4. Authentication: Clerk handles dev/prod environments automatically

• Python 3.9+ installed
• Node.js 18+ installed
• MongoDB Atlas account created
• Clerk account setup with project created
• Environment variables configured
• Dependencies installed
• Both servers running
• Application accessible at localhost:5173

• Frontend Application: https://summary.techycsr.dev
• Backend API: https://summaryapi.techycsr.me
• Health Endpoint: https://summaryapi.techycsr.dev/api/v1/health

GET  /api/v1/health # System health check
POST  /api/v1/upload-resume # Document upload & analysis
GET  /api/v1/insights # User document history
GET  /api/v1/document/{id}/preview # PDF preview (authenticated)

{
 "status": "healthy",
 "timestamp": "2025年08月30日T13:27:21.331972",
 "environment": "production",
 "database": {
 "connected": true,
 "status": "connected",
 "error": null,
 "insights_count": 42
 }
}

POST /api/v1/upload-resume
Content-Type: multipart/form-data
Authorization: Bearer {jwt_token}
file: {pdf_file}
provider: "sarvam" | "gemini"

{
 "summary": "**👤 Name:** John Doe\n**📧 Contact:** john.doe@email.com, +1-555-0123\n**💼 Professional Summary:** Experienced software engineer with 5+ years...",
 "provider": "sarvam",
 "is_fallback": false,
 "filename": "resume.pdf",
 "upload_date": "2025年08月30日T13:27:21.331972",
 "document_id": "66d1b2c3d4e5f6789abcdef0"
}

👤 Name: [Extracted full name]
📧 Contact: [Phone, Email, Location]
💼 Professional Summary: [Key qualifications highlights]
🎯 Core Skills: [Technical and soft skills]
💪 Experience Highlights:
 • [Most relevant role with quantified achievements]
 • [Second important position with metrics]
 • [Third significant role with impact data]
🎓 Education: [Degrees, institutions, GPA if available]
🏆 Notable Achievements:
 • [Top accomplishment with quantified results]
 • [Second significant achievement]
 • [Third notable accomplishment]
📊 Career Insights:
 • Years of Experience: [Calculated total]
 • Industry Focus: [Primary domain]
 • Career Level: [Entry/Mid/Senior/Executive]

📄 Document Type: [Auto-detected type]
📝 Document Summary: [Comprehensive overview]
🔍 Key Insights: [Major findings and observations]
📊 Main Topics: [Primary and secondary topics]
💡 Critical Information: [Important facts and recommendations]
🎯 Target Audience: [Intended readers]
📈 Key Takeaways: [Actionable insights]

The system automatically detects document types:

• Resume/CV: Professional experience documents
• Research Paper: Academic and scientific documents
• Proposal: Project and business proposals
• Legal Document: Contracts and agreements
• Report: Analysis and findings documents
• General Document: Other document types

• File Validation: PDF-only, 10MB size limit
• Text Extraction: Advanced PDFPlumber integration
• Content Sanitization: Clean text preprocessing
• Progress Tracking: Real-time upload status

• Document Type Detection: Intelligent classification
• Dual AI Integration: Primary (Sarvam) + Secondary (Gemini)
• Intelligent Fallback: Keyword frequency analysis
• Structured Output: Formatted, actionable insights

• Secure Authentication: Clerk-based JWT validation
• Personal History: Complete analysis tracking
• PDF Preview: Authenticated document viewing
• Session Management: Persistent user sessions

• Responsive Design: Mobile detection with desktop optimization
• Error Handling: Professional 404 pages and error management
• Health Monitoring: System status tracking
• Performance Optimization: Async processing and caching

• Desktop Optimized: Full-featured dashboard experience
• Mobile Detection: Automatic redirection to mobile-optimized messaging
• Tablet Support: Warning banners for limited mobile functionality
• Progressive Enhancement: Graceful degradation across devices

• Modern Dashboard: Clean, professional layout
• Drag-and-Drop Upload: Intuitive file handling
• Real-time Feedback: Progress indicators and status updates
• Theme Support: Dark/light mode toggle
• Error Recovery: User-friendly error messages and recovery options

• Serverless Architecture: Auto-scaling Vercel functions
• Async Processing: Non-blocking I/O operations
• Connection Pooling: Optimized MongoDB connections
• Error Recovery: Graceful degradation and reconnection logic

• Code Splitting: Dynamic imports for reduced bundle size
• Lazy Loading: On-demand component loading
• Caching Strategy: Optimized API response caching
• Bundle Analysis: Size optimization and tree shaking

• Indexed Queries: Optimized user-based data retrieval
• Document Storage: Efficient binary PDF storage
• Connection Management: Serverless-optimized pooling

• JWT Validation: Secure token-based authentication
• User Isolation: Strict data access controls
• Session Management: Secure session handling
• API Protection: Authenticated endpoint access

• File Validation: Strict PDF-only upload enforcement
• Size Limits: 10MB maximum file size
• Content Sanitization: Safe text processing
• Secure Storage: Encrypted MongoDB Atlas storage

• HTTPS Enforcement: SSL/TLS encryption
• Environment Variables: Secure configuration management
• API Rate Limiting: Protection against abuse
• Error Handling: Secure error message disclosure

✅ Backend API: Operational (99.9% uptime)
✅ Frontend App: Deployed & Responsive
✅ Database: MongoDB Atlas Connected (42 documents)
✅ AI Services: Sarvam + Gemini Operational
✅ Authentication: Clerk Integration Active
✅ File Processing: PDF Upload & Analysis Working
✅ Mobile Support: Detection & Redirection Active
✅ Error Handling: 404 Pages & Recovery Implemented

• Production-Ready: Fully deployed and operational system
• Enterprise-Grade: Professional UI/UX with comprehensive error handling
• AI-Specialized: Optimized for resume analysis with fallback intelligence
• Scalable Architecture: Serverless deployment with auto-scaling capabilities
• Modern Tech Stack: Latest frameworks and best practices implementation
• Security-First: Comprehensive authentication and data protection
• Performance-Optimized: Fast loading times and efficient processing

Built with modern web technologies for professional document analysis workflows. Deployed and operational at summary.techycsr.me "filename": "resume.pdf", "upload_date": "2025-08-28T14:00:00Z", "provider": "sarvam", "summary": "Professional summary...", "is_fallback": false, "file_size": 1234567 } ], "total_count": 1 }

## 🚀 Deployment
### Prerequisites for Deployment
1. **Vercel Account** - For hosting both frontend and backend
2. **MongoDB Atlas** - Cloud database
3. **Clerk Account** - Authentication service
4. **Domain** (optional) - For custom domain
### Backend Deployment (Vercel)
1. **Connect to Vercel**
```bash
cd backend
npm i -g vercel # Install Vercel CLI
vercel # Follow the prompts

2. Set Environment Variables Go to your Vercel dashboard and add:

• MONGODB_URI
• CLERK_SECRET_KEY
• SARVAM_API_KEY
• GEMINI_API_KEY
• ENVIRONMENT=production
• DEBUG=False
• ALLOWED_ORIGINS=https://your-frontend-domain.vercel.app

1. Connect to Vercel

cd frontend
vercel # Follow the prompts

2. Set Environment Variables Add to Vercel dashboard:

• VITE_CLERK_PUBLISHABLE_KEY
• VITE_API_BASE_URL=https://your-backend-domain.vercel.app/api/v1

1. Update Clerk Settings

   • Add your production domains to allowed origins
   • Update redirect URLs

2. Update API CORS

   • Add production frontend URL to backend CORS settings

3. Test the Deployment

   • Verify authentication flow
   • Test file upload functionality
   • Check AI provider integration

Built with ❤️ by @TechyCSR

Professional full-stack developer specializing in AI-powered applications and modern web technologies.

© 2025 TechyCSR • AI-Powered Document Analysis Platform • summary.techycsr.me