RFQ Processing Platform

An AI-powered platform for automating supplier matching in procurement processes, with specialized support for AI hardware requirements.

🚀 Features

• Intelligent RFQ Processing: Upload RFQ documents (PDF/text) and extract requirements using AI
• Smart Supplier Matching: Match suppliers based on technical specifications and business criteria
• AI Hardware Specialization: Advanced matching for GPUs, accelerators, and ML hardware
• Compliance Checking: Automated export control and regulatory compliance verification
• Email Automation: Generate and send professional supplier proposals

🏗 Architecture

The platform consists of:

• Frontend: React/TypeScript client with modern UI components
• Backend: Python FastAPI server with AI services
• Database: PostgreSQL with Drizzle ORM
• AI Services: Integration with OpenAI/Featherless AI for NLP tasks

🛠 Setup & Installation

Prerequisites

• Node.js 20+
• Python 3.11+
• PostgreSQL database

Environment Variables

Create a .env file in the root directory:

DATABASE_URL=your_postgresql_connection_string
OPENAI_API_KEY=your_openai_api_key  # For embeddings and AI processing
FEATHERLESS_API_KEY=your_featherless_key  # For requirement extraction

Quick Start

1. Install dependencies:

npm install

2. Start the application:

npm run dev

The application will be available at:

• Frontend: http://localhost:5000
• API: http://localhost:8000

📊 Usage

Basic Workflow

1. Upload RFQ: Submit procurement documents via the upload interface
2. Review Requirements: AI extracts and categorizes technical specifications
3. Match Suppliers: System finds relevant suppliers based on capabilities
4. Generate Proposals: Create professional email proposals for selected suppliers

AI Hardware Platform

For specialized AI/ML hardware procurement:

1. Use the questionnaire interface for technical requirements
2. System matches GPUs, accelerators, and compute infrastructure
3. Includes compliance checking for export controls
4. Performance benchmarking and framework compatibility

🔧 API Endpoints

Core Endpoints

• GET /api/rfqs - List all RFQs
• POST /api/rfqs/upload - Upload RFQ document
• POST /api/rfqs - Create RFQ manually
• POST /api/rfqs/{id}/match-suppliers - Find supplier matches
• POST /api/proposals/{id}/generate-email - Generate email proposals

AI Hardware Endpoints

• POST /api/seed-ai-hardware-products - Initialize sample data
• GET /api/ai-hardware/check-compliance - Compliance verification
• GET /api/ai-hardware/frameworks-compatibility - Framework support check
• GET /api/ai-hardware/performance-comparison - Hardware benchmarking

🧪 Development

Frontend Development

cd client
npm run dev

Backend Development

cd python_backend
python -m uvicorn api.app:app --reload --port 8000

Database Schema

npm run db:push

📦 Deployment

Deploy on Replit:

1. Import project to Replit
2. Configure environment variables in Replit Secrets
3. The project will auto-deploy with the configured run command

🚧 Known Limitations (MVP)

• Email sending is simulated (not actually sent)
• Limited supplier database (sample data)
• Basic compliance checking (not comprehensive)
• Single-user system (no authentication)

🔮 Roadmap

• Real email integration (SendGrid/SMTP)
• Advanced supplier database with real integrations
• Multi-user support with authentication
• Advanced analytics and reporting
• Mobile-responsive design improvements

📄 License

MIT License - see LICENSE file for details