Finchat

📖 Overview

A RAG (Retrieval Augmented Generation) based chat application capable of providing realtime stock quotes, market insights, stock news, historical earnings as well as creating a meaningful visualization of that data on the fly.

FinChat.UI.Walkthrough.mp4

📚 Stack

Frontend: Streamlit, Pandas

Backend: FastAPI, Docker, FinnHub

AI/LLM: LangChain, LangGraph, Azure OpenAI

🏛️ App Architecture

Three core components work together:

1. client.py - Streamlit frontend with chat interface
2. server.py - FastAPI backend handling AI processing
3. llm.py - LangGraph workflow with financial data tools

✨ Features

• Real-time stock data analysis
• Company recommendation trends visualization
• Earnings history and news summaries
• Conversational AI with financial expertise
• Dockerized application for easy deployment

🛠️ Local Setup

Prerequisites

• Docker and Docker Compose
• API keys for Finnhub and Azure OpenAI

Installation

1. Clone the repository
2. Create a .env file in the root directory with the following content:

OPENAI_API_DEPLOYMENT=your-deployment-name
OPENAI_API_MODEL=your-model-name
AZURE_OPENAI_ENDPOINT=your-azure-endpoint
OPENAI_API_VERSION=2023-05-15
OPENAI_API_KEY=your-openai-key
FINNHUB_API_KEY=your-finnhub-key

3. Build and run the Docker containers:

docker-compose up --build

🚀 Usage

After running the Docker containers:

1. Access the Streamlit frontend at http://localhost:8502

2. The FastAPI backend will be available at http://localhost:8000

3. Sample Queries:

Q. "Show me recommendation trends for apple"
Q. "What's the current price of tesla?"
Q. "Summarize recent news for microsoft"
Q. "Display earnings history for google"

🧩 Component Interaction

1. Client (Streamlit Frontend)

• Handles user interface and chat history
• Sends prompts to server via POST requests
• Visualizes responses using Streamlit charts
• Maintains session-based chat history

2. Server (FastAPI Backend)

• Receives POST requests with user prompts
• Maintains conversation state using LangGraph
• Coordinates with financial data tools
• Returns AI-generated responses in JSON format

3. LLM Workflow (LangGraph)

• Processes natural language queries using Azure OpenAI
• Routes to appropriate financial tools:
  • getStockData: Company profiles
  • getStockRecommendation: Analyst trends
  • getCompanyNews: Recent news summaries
  • getStockPrice: Real-time quotes
  • getCompanyEarnings: Historical performance

🐳 Docker Configuration

The application is containerized using Docker for easy deployment and consistency across environments.

1. Dockerfile: Defines the environment for both the server and client.
2. docker-compose.yml: Orchestrates the multi-container setup:
   • server: Runs the FastAPI backend
   • client: Runs the Streamlit frontend

To modify ports or environment variables, adjust the docker-compose.yml file.

📄 License

MIT License - Use responsibly with proper API key management. Always verify financial insights with professional advisors.