FinSight.ai - Your AI-Powered Financial Analyst

Submission for the Hack-Nation Global AI Hackathon (Aug 2025)

---

1. Project Description

FinSight.ai is a full-stack, AI-powered application designed to solve the problem of information overload in financial analysis. It transforms dense, impenetrable SEC 10-K filings and real-time market news into a clear, actionable, and interactive investment briefing.

Our system allows a user to enter any major US stock ticker and receive a comprehensive, AI-generated report in seconds. This report includes a strategic executive summary, a data-driven Bull vs. Bear case, a snapshot of key financial metrics, and a summary of critical risks admitted to by the company itself.

The core innovation is a stateful, conversational RAG (Retrieval-Augmented Generation) pipeline. After viewing the report, users can engage in an interactive Q&A to interrogate the source documents directly, with every answer grounded in verifiable evidence, making our tool not just intelligent, but trustworthy.

2. Tech Stack

• Backend: Python, FastAPI, LangGraph (for agent orchestration)
• Frontend: Next.js, React, Tailwind CSS
• AI / ML Models:
  • Synthesis: Gemini 1.5 Pro
  • Chat & Routing: Gemini 1.5 Flash
  • Embeddings: Google embedding-001
• Vector Store: FAISS (in-memory)
• Data APIs: SEC-API, Tavily Search, yfinance

3. How to Run Locally

This is a full-stack application with a Python backend and a Next.js frontend. You will need two terminals running simultaneously from this root project folder.

Prerequisites

• Python 3.10+
• Node.js and npm
• API keys (see .env.example)

Terminal 1: Start the Python Backend

1. Set up the environment:

   # Create and activate a virtual environment (recommended)
   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   
   # Install Python dependencies
   pip install -r requirements.txt

2. Add API Keys:

   • Create a file named .env in the root directory.
   • Copy the contents of .env.example into it and fill in your actual API keys.

3. Run the server:

   uvicorn main:api_app --reload

   The backend will be running at http://127.0.0.1:8000.

Terminal 2: Start the Next.js Frontend

1. Set up the environment:

   # Install Node.js dependencies
   npm install --legacy-peer-deps

2. Add API URL:

   • Create a file named .env.local in the root directory.
   • Add the following line to it:

     NEXT_PUBLIC_API_BASE_URL=http://127.0.0.1:8000
     

3. Run the server:

   npm run dev

   The frontend will be running at http://localhost:3000.

4. Open http://localhost:3000 in your browser to use the application.