VentureMatch AI ✨

Where Visionaries find Capital through deep semantic alignment.

VentureMatch AI is a production-level RAG (Retrieval-Augmented Generation) platform designed to connect founders and investors. Unlike traditional keyword-based matching, VentureMatch uses high-dimensional vector embeddings to understand the true "thesis" of an investor and the "vision" of a startup.

🚀 Key Features

• AI Document Extraction: Automatically parses pitch decks (PDF/TXT) using Google Gemini 2.5 Flash to build high-precision structured profiles.

• Deterministic Chat Loop: A smart interactive UI powered by Gemini's reasoning to identify missing data in your profile and ask context-aware questions.

• Semantic Matchmaking Engine: Powered by ChromaDB and Gemini Multimodal Embeddings to find best-fit matches based on actual content similarity.

• Authenticated Dashboards: Separate secure portals for Founders and Investors (Node.js/Express + MongoDB).

• Conversational RAG Chat: Chat directly with matched profiles using Gemini 1.5/2.5 to ask deep questions about business models or investment theses.

• Frontend: React (Vite), Vanilla CSS (Glassmorphism), Lucide Icons.

• Auth Backend: Node.js, Express, MongoDB (Mongoose), JWT.

• AI Backend: Python (FastAPI), Google Gemini 2.5 Flash, ChromaDB, LangChain, PyPDF2.

🛠️ Built With

VentureMatch AI leverages a modern, distributed architecture to handle intensive AI operations:

• Languages: Python (Backend AI), JavaScript (Frontend & Auth), CSS3 (Modern Styling).
• Frontend Framework: React.js via Vite for high-performance builds.
• AI Engine: Google Gemini 2.5 Flash API for multimodal extraction and conversational reasoning.
• Vector Database: ChromaDB for lightning-fast semantic similarity search.
• Primary Backends:
  • FastAPI (Python) for asynchronous AI service orchestration.
  • Express.js (Node.js) for robust user authentication and session management.
• Database: MongoDB for persistent storage of user profiles and authentication data.
• State Management: React Context API & LocalStorage for cross-platform data synchronization.

📦 Installation & Setup

1. Clone the Repository

git clone https://github.com/djivites/ventureMatch.git
cd ventureMatch

2. Configure Environment Variables

Create .env files in backend/ and auth-backend/ as per the templates provided in the documentation.

3. Run the Backend (Python)

cd backend
python -m venv venv
# Activate venv
pip install -r requirements.txt
python -m uvicorn main:app --reload

4. Run the Auth Server (Node)

cd auth-backend
npm install
npm run dev

5. Run the Frontend (React)

cd frontend
npm install
npm run dev

📄 License

MIT License