MarketBridge – AI-Driven Marketing Campaign Planner (NHCE Silver Jublee TechFest Hackathon 2025 Winning Project!)

MarketBridge is an intelligent multi-agent platform that automates marketing campaign planning by connecting marketing, finance, and inventory teams through autonomous AI agents.

The system uses Retrieval-Augmented Generation (RAG) to access CRM, finance, and inventory data in real time, allowing businesses to plan, optimize, and adapt campaigns more efficiently.

---

Overview

Marketing campaign planning typically requires coordination between multiple departments:

• Marketing: Responsible for creative ideas and targeting customer segments.
• Finance: Ensures financial feasibility, budget control, and ROI.
• Inventory: Confirms product availability and supply chain readiness.

MarketBridge simplifies this process by using AI agents that collaborate and negotiate automatically to produce data-driven campaign plans.

---

Features

• Multi-Agent Collaboration The system includes:

  • Creative Agent – Generates campaign ideas, creatives, and customer segments.
  • Finance Agent – Evaluates financial feasibility, budgets, and profitability.
  • Inventory Agent – Monitors product availability and distribution readiness.

• Adaptive Planning Campaigns automatically adjust when inventory levels or budgets change.

• RAG-Enabled Intelligence Integrates real-time CRM, finance, and inventory data to make informed decisions.

• What-If Simulation Allows users to test scenarios such as discount variations or budget changes.

• Modern UI/UX A clean, minimal interface design style.

• Interactive Dashboards Displays agent interactions, campaign projections, and overall performance.

---

Tech Stack

Layer	Technologies
Frontend	React 18, Vite, Framer Motion
Backend	Python 3, Flask or FastAPI
AI / LLM	Google Generative AI (Gemini API)
Data Handling	Retrieval-Augmented Generation (RAG)
Deployment	Vercel, Render, or Railway (optional)

---

Installation and Setup

Frontend Setup

1. Clone the repository:

   git clone https://github.com/Veins19/MarketBridge.git
   cd MarketBridge/frontend

2. Install dependencies:

   npm install

3. Run the development server:

   npm run dev

4. Open your browser and navigate to http://localhost:5173

---

Backend Setup

1. Move to the backend directory and create a virtual environment:

   cd ..
   /backend
   python3 -m venv venv
   source venv/bin/activate

2. Install dependencies:

   pip install -r requirements.txt

3. Run the backend server:

   python main.py

4. The backend will run on http://localhost:5000

---

Environment Variables

Create .env files in both the frontend and backend directories.

Frontend .env

VITE_API_URL=http://localhost:5000

Backend .env

GOOGLE_API_KEY=your_google_generative_ai_key

---

Folder Structure

MarketBridge/
│
├── frontend/
│   ├── src/
│   │   ├── components/     # UI components
│   │   ├── pages/          # Pages such as Home, Dashboard
│   │   ├── styles/         # Global CSS 
│   │   ├── App.jsx         # Main app routes
│   │   └── main.jsx        # Vite entry file
│   └── package.json
│
├── backend/
│   ├── agents/             # Creative, Finance, and Inventory agents
│   ├── agent_manager.py    # Orchestrates agent collaboration
│   ├── main.py             # Backend entry point
│   ├── requirements.txt
│   └── ...
│
└── README.md

---

Agent Workflow

1. The Creative Agent proposes marketing ideas, content, and segments.
2. The Finance Agent validates these plans based on budget and ROI.
3. The Inventory Agent checks stock levels and supply constraints.
4. Agents collaborate and negotiate to finalize a feasible, optimized campaign plan.
5. The resulting plan is displayed to the user along with supporting data and visualizations.

---

UI and UX Highlights

• Clean, minimal landing page with clear CTAs
• Dashboard layout featuring cards for each agent
• Smooth transitions and animations with Framer Motion
• Fully responsive design for all devices
• Consistent color palette and modern typography inspired by Quiccle.com

---

Future Enhancements

• Real-time agent collaboration using WebSockets
• Campaign analytics dashboard for deeper insights
• Visualization of agent conversations and reasoning
• Budget optimization module
• User authentication and role-based permissions

---

Troubleshooting

Issue	Solution
npm: command not found	Install Node.js using brew install node
ModuleNotFoundError: packaging	Run pip install packaging inside the backend venv
CORS error	Add flask-cors or FastAPI CORSMiddleware
Port already in use	Kill running process or change port in vite.config.js
Blank page	Check .env variables and ensure the API URL is correct

---

Contributing

1. Fork the repository

2. Create a new branch:

   git checkout -b feature/new-ui

3. Commit your changes:

   git commit -m "Improved dashboard design"

4. Push the branch:

   git push origin feature/new-ui

5. Open a Pull Request

---

License

This project is licensed under the MIT License, allowing free use, modification, and distribution.

---

Summary

Category	Description
Purpose	AI-driven marketing campaign automation
Frontend	React + Vite + Framer Motion
Backend	Python (Flask/FastAPI) with multi-agent logic
AI Engine	Google Generative AI (Gemini API)
Design	Clean, minimal interface

---