RFP AI-Powered Application

An intelligent Request for Proposal (RFP) platform that connects buyers with vendors using AI-powered matching and automated workflows.

Live Link: https://rfp-connect.vercel.app/

🚀 Features

AI-Powered RFP Processing: Automated parsing and analysis of RFP requirements using OpenAI
Smart Vendor Matching: Intelligent algorithm to match RFPs with suitable vendors
Real-time Notifications: Email notifications for RFP updates and matches
User Authentication: Secure JWT-based authentication for buyers and vendors
Background Job Processing: Redis-powered queue system for async operations
Responsive UI: Modern React frontend with Tailwind CSS and Framer Motion

🏗️ System Architecture

High-Level Architecture

Frontend (Vercel)

📱 React + TS
🎨 Tailwind CSS
⚛️ Redux Toolkit

Backend (Render)

🚀 Express API
🔐 JWT Auth
🧠 OpenAI GPT

Services

🗄️ PostgreSQL
🟥 Redis Queue
📧 SendGrid

Frontend (React + TypeScript)

Framework: React 19 with TypeScript
Styling: Tailwind CSS with custom animations
State Management: Redux Toolkit
Routing: React Router DOM
UI Components: Custom components with Framer Motion animations

Backend (Node.js + Express)

Runtime: Node.js with ES modules
Framework: Express.js
Database: PostgreSQL with Sequelize ORM
Queue System: BullMQ with Redis
AI Integration: OpenAI API for RFP analysis
Email Service: Nodemailer with SendGrid

🗄️ Database Schema & Relationships

Entity Relationship Diagram

Table	Key Fields	Relationships
Buyers	id(UUID PK), email(unique), phone(unique), status	1:N → RFPs
Vendors	id(UUID PK), companyName, mainService, status	1:N → VendorProductsM:N → RFPs
RFPs	id(UUID PK), buyerId(FK), title, aiParsed(JSON), status	N:1 ← BuyersM:N → Vendors
RfpVendors	rfpId(PK), vendorId(PK), score(0-100), accepted	Junction table
VendorProducts	id(PK), vendorId(FK), name, price, category	N:1 ← Vendors

Table Relationships & Cardinality

Core Entities

Buyers (1:N with RFPs)
- One buyer can create multiple RFPs
- Each RFP belongs to exactly one buyer
Vendors (1:N with VendorProducts)
- One vendor can have multiple products/services
- Each product belongs to exactly one vendor
RFPs (M:N with Vendors through RfpVendors)
- One RFP can be matched with multiple vendors
- One vendor can be matched with multiple RFPs

Junction Tables

RfpVendors (Many-to-Many Resolution)
- Composite Primary Key: (rfpId, vendorId)
- Stores matching score, notification status, proposal acceptance
- Links buyers, RFPs, and vendors for the matching process
EmailVerificationTokens (1:N with Users)
- Supports both buyer and vendor email verification
- Time-limited tokens with expiration tracking

Database Models Detail

Buyers Table

- id (UUID, PK)
- firstName, lastName (STRING, NOT NULL)
- emailAddress (STRING, UNIQUE, NOT NULL)
- phone (STRING, UNIQUE, NOT NULL)
- address, city, state, country, postalCode (STRING)
- passwordHash (STRING, NOT NULL)
- isEmailVerified (BOOLEAN, DEFAULT false)
- emailVerifiedAt (DATE)
- status (ENUM: PENDING_VERIFICATION, ACTIVE, SUSPENDED)

Vendors Table

- id (UUID, PK)
- vendorCompanyName (STRING, NOT NULL)
- emailAddress (STRING, UNIQUE, NOT NULL)
- mainProductService (STRING)
- passwordHash (STRING, NOT NULL)
- phoneNumber, address, pincode, state (STRING)
- establishedDate (DATE)
- deliveryTimeframe (STRING)
- isEmailVerified (BOOLEAN, DEFAULT false)
- status (ENUM: PENDING_VERIFICATION, ACTIVE, SUSPENDED)
- onboardingCompleted (BOOLEAN, DEFAULT false)
- onboardingProgress (INTEGER, DEFAULT 0)

RFPs Table

- id (UUID, PK)
- buyerId (UUID, FK → Buyers.id)
- rfpTitle (STRING, NOT NULL)
- buyerExpectations (TEXT, NOT NULL)
- expectedDelivery (STRING, NOT NULL)
- budgetRange (STRING)
- additionalContext (TEXT)
- aiParsedRequirements (JSON) -- AI analysis results
- shortlistedVendorsList (JSON) -- Array of vendor IDs
- rfpStatus (ENUM: Pending, Parsed, Shortlisted, Notified)
- notificationSent (BOOLEAN, DEFAULT false)
- submissionDate (DATE, DEFAULT NOW)

RfpVendors Junction Table

- rfpId (UUID, PK, FK → RFPs.id)
- vendorId (UUID, PK, FK → Vendors.id)
- buyerId (UUID, FK → Buyers.id)
- vendorScore (INTEGER, DEFAULT 0) -- Matching algorithm score
- notificationSent (BOOLEAN, DEFAULT false)
- proposalAccepted (BOOLEAN, NULL) -- Buyer's decision

🤖 AI-Powered Workflow

RFP Submission Flow Buyer submits RFP → Queue Job → AI Analysis → Vendor Matching → Notifications
AI Processing Pipeline Natural Language Processing: OpenAI GPT-4 analyzes RFP requirements

Structured Extraction: Converts free text to structured data:

{
  "summary": "Brief RFP description",
  "businessDomain": "e.g., SaaS, e-commerce",
  "primaryGoal": "Main objective",
  "techStack": ["React", "Node.js", "PostgreSQL"],
  "mustHaveFeatures": ["Authentication", "API"],
  "niceToHaveFeatures": ["Analytics", "Mobile"],
  "expectedDelivery": "Cleaned timeline"
}

Vendor Matching Algorithm

Scoring System (0-100 points):

Tech Stack Match (0-15 pts): 3 pts per matching technology
Feature Match (0-10 pts): 2 pts per required feature
Location Proximity (0-5 pts): Same state/region preference
Delivery Timeline (0-2 pts): Vendor delivery capability
Product Portfolio (0-5 pts): 0.5 pts per relevant product
Onboarding Status (0-3 pts): Completed vendor profiles

Minimum Score: 8 points for shortlisting
Maximum Results: Top 5 vendors per RFP

🔄 Background Job Processing

Queue System (BullMQ + Redis)

RFP Processing Job

Triggered on RFP submission
AI parsing of requirements
Vendor matching algorithm execution
Email notification dispatch

Job Flow

RFP Submit → Queue Job → AI Parse → Update RFP → Match Vendors → 
Create RfpVendor Records → Send Notifications → Update Status

📋 Prerequisites

Node.js (v18 or higher)
Docker & Docker Compose
PostgreSQL (if running locally)
Redis (if running locally)

🛠️ Installation & Setup

Clone the Repository

git clone <repository-url>
cd rfp-ai-powered-app

2. Environment Configuration
Backend Environment (.env in /backend)
PORT=8080
FE_PORT=3000
PROTOCOL=http
DB_USER=postgres
DB_PASSWORD=Test1234
DB_HOST=localhost
DB_NAME=ai_powered_rfp_db
DB_PORT=5432
JWT_SECRET_KEY=your_secret_key_here
EMAIL_USER=your_email@gmail.com
SENDGRID_API_KEY=your_sendgrid_api_key
OPENAI_SECRET_KEY=your_openai_api_key
REDIS_HOST=localhost
REDIS_PORT=6379

Frontend Environment (.env in root)

REACT_APP_API_URL=http://localhost:8080

Using Docker (Recommended)

# Start all services
docker-compose up -d

# View logs
docker-compose logs -f

# Stop services
docker-compose down

4. Manual Setup
Backend Setup
cd backend
npm install
npm run dev

Frontend Setup
npm install
npm start

Database Setup
# Start PostgreSQL and Redis
docker-compose up -d db redis

# Database will be automatically created by Sequelize

🌐 API Endpoints
Authentication
POST /api/auth/login - User login

POST /api/auth/register - User registration

Buyers
GET /api/buyer - Get buyer profile

POST /api/buyer - Create buyer profile

PUT /api/buyer - Update buyer profile

Vendors
GET /api/vendor - Get vendor profile

POST /api/vendor - Create vendor profile

PUT /api/vendor - Update vendor profile

RFPs
POST /api/rfp - Submit new RFP

GET /api/rfp - Get RFPs (filtered by user role)

Email Verification
POST /api/verify-email - Verify email address

🎨 Frontend Structure src/
├── components/
│ ├── pages/ # Page components
│ └── ui/ # Reusable UI components
├── context/ # React contexts
├── entities/ # TypeScript type definitions
├── guards/ # Route protection
├── services/ # API service layer
├── store/ # Redux store configuration
└── staticData/ # Static content

🔧 Development Available Scripts Frontend npm start # Start development server npm run build # Build for production npm test # Run tests

Backend
npm start          # Start production server
npm run dev        # Start development server with nodemon

Code Style

ESLint configuration for consistent code style
TypeScript for type safety
Modular architecture with separation of concerns

🚀 Deployment Production Build

Build frontend

npm run build

Start backend in production

cd backend && npm start

Docker Production
docker-compose -f docker-compose.prod.yml up -d

🤝 Contributing

Fork the repository

Create a feature branch (git checkout -b feature/amazing-feature)

Commit your changes (git commit -m 'Add amazing feature')

Push to the branch (git push origin feature/amazing-feature)

Open a Pull Request

📝 License This project is licensed under the MIT License - see the LICENSE file for details.

🆘 Support

For support and questions:

Create an issue in the GitHub repository
Contact the development team

🔮 Future Enhancements

Advanced AI matching algorithms
Real-time chat between buyers and vendors
Document management system
Advanced analytics dashboard
Mobile application
Multi-language support

Name		Name	Last commit message	Last commit date
Latest commit History 18 Commits
backend		backend
public		public
src		src
.dockerignore		.dockerignore
.gitignore		.gitignore
Backend table details.txt		Backend table details.txt
Dockerfile		Dockerfile
Makefile		Makefile
README.md		README.md
docker-compose.yml		docker-compose.yml
package-lock.json		package-lock.json
package.json		package.json
postcss.config.js		postcss.config.js
tailwind.config.js		tailwind.config.js
tsconfig.json		tsconfig.json

SAURABH200301/RFP_CONNECT

Folders and files

Latest commit

History

Repository files navigation