💸 FinGPT - AI-Powered Financial Assistant

Team Sniders

---

🔍 Introduction

India is witnessing a surge of new investors—millions stepping into the world of finance. Yet, financial literacy remains alarmingly low. Manual advisory isn’t scalable, and most platforms only cater to seasoned investors. That’s where FinGPT steps in — an intelligent, accessible, and AI-powered financial assistant designed for everyone.

---

🌟 Unique Selling Proposition (USP)

✅ 100% Free & Scalable – Powered by Google tools, no fees or hidden charges.

✅ Real-Time Insights – Instant stock data from Yahoo Finance & Google Finance.

✅ Multi-Language Support – Chat in any Indian language via Gemini AI.

✅ Privacy-Focused – No personal data stored beyond chat history.

✅ Personalized Investment Suggestions – Smart AI-driven recommendations.

✅ Jargon-Free Learning – Simplified finance for absolute beginners.

---

❓ Problem Statement

🚫 New investors struggle with:

• Low financial literacy
• Confusing jargon
• Lack of personalized support
• Overwhelming or technical platforms

💡 FinGPT aims to make finance simple, smart, and accessible to the masses.

---

🛠️ Solution Overview

FinGPT is a GenAI-Powered Financial Assistant that:

💬 Answers investment questions in simple, conversational language
📉 Shows real-time stock data and interactive charts
📈 Predicts trends with live buy/sell suggestions
🔒 Keeps user data private & secure
📚 Educates users without technical jargon

---

🧠 Features & Capabilities

✨ Built on top of Google Gemini AI, FinGPT can:

🔹 Classify user queries into investment-related intents

🔹 Provide live stock prices, graphs, and charts

🔹 Generate stock predictions and personalized tips

🔹 Send friendly, beautified replies for a smooth user experience

---

🧰 Technology Stack

🧩 Component	🔧 Tool/Technology Used	💡 Purpose
Conversational AI	Google Gemini + LangChain	Smart finance chat assistant
NLP & Intent Classification	Fine-tuned Google Gemini + LangChain	Handling chat flows and intents
Chat Interface	React + TypeScript	Frontend chatbot interface
Data Storage	MongoDB	User history & preferences
Market Insights	Yahoo Finance API	Live stock & mutual fund data
Visualization & Charts	Google Sheets + Plotly	Interactive charts & graphs
Financial Data Analysis	Python + Google Colab	AI model training and data crunching
Realtime Analytics	Google Analytics	Track engagement and visualize performance
APIs Integration	FastAPI	Custom endpoints for NLP Chatbot

---

📊 SWOT Analysis

🟩 Strengths	🟥 Weaknesses
✅ Fully AI-powered and automated financial assistant ✅ Cost-effective with free, serverless tools (Gemini) ✅ Real-time market insights and personalized recommendations	⚠️ Dependent on Google’s ecosystem – limited customization ⚠️ No direct trading functionality – purely advisory
🟨 Opportunities	🟦 Threats
🚀 Massive market of untapped new investors in need of financial guidance 📱 Expansion to WhatsApp, Google Assistant, and voice platforms 💼 Potential monetization via premium insights and advisory tiers	🧠 Competition from large fintechs with integrated trading & advisory platforms ⚖️ Regulatory shifts in financial advisory laws

---

📁 Project Structure

├── backend/                     # Backend for managing APIs and logic
│   ├── src/                     # Backend source files
│   ├── package.json             # Backend dependencies
│   ├── tsconfig.json            # TypeScript config for backend

├── backend_ai/                  # AI engine for Gemini-based financial assistant
│   ├── app.py                   # Main Flask/FastAPI application entry point
│   ├── model.py                 # AI model logic and setup
│   ├── model_config.py          # Configuration for model parameters
│   ├── requirements.txt         # Python package dependencies
│   ├── tools.py                 # Utility tools for AI processing
│   └── .gitignore               # Git ignore rules for backend_ai

├── frontend/                    # Frontend UI of the application
│   ├── public/                  # Static assets (icons, images, etc.)
│   ├── src/                     # Frontend source code (TypeScript)
│   ├── .gitignore               # Git ignore rules for frontend
│   ├── README.md                # Frontend-specific README
│   ├── eslint.config.js         # ESLint configuration
│   ├── index.html               # Root HTML file
│   ├── package.json             # Frontend dependencies
│   ├── tsconfig.app.json        # TypeScript config for app
│   ├── tsconfig.json            # General TypeScript config
│   ├── tsconfig.node.json       # TypeScript config for node scripts
│   └── vite.config.ts           # Vite config for development server

├── graph/                       # Data visualization and stock graph generation
│   ├── graph.py                 # Script for plotting live charts and predictions
│   └── .gitignore               # Git ignore rules for graph module

├── GenAI-Powered Financial Assistant.pdf   # Project documentation (PDF)
├── README.md                                # Main project documentation

🚀 How to Run

1. Clone the repository:

git clone https://github.com/SagnikBasak04/finance_bot.git
cd finance_bot

2. Run the AI Agent server:

cd backend_ai
pip install -r requirements.txt
python app.py

3. Run the Graph server:

cd ../graph
python graph.py

4. Run the backend server:

cd ../backend
npm install
npm run dev

5. Run the frontend server:

cd ../frontend
npm install
npm run dev

---

👨‍💻 Contributors

👤 Name	💼 Role	🔗 GitHub
Tamojit Das (Team Lead)	Full-Stack Developer	GitHub
Sagnik Basak	AI Engineer & Data Analyst	GitHub
Anidipta Pal	AI & Backend Developer	GitHub
Titas Kabiraj	UI-UX & Frontend Developer	GitHub