Skip to content

README.md

Latest commit

 

History

History
168 lines (108 loc) · 4.94 KB

README.md

File metadata and controls

168 lines (108 loc) · 4.94 KB

CorpMindAI 🧠

CorpMindAI is a secure, intelligent corporate knowledge base that allows teams to chat with their documents. By leveraging RAG (Retrieval-Augmented Generation) technology, it provides accurate, context-aware answers based on uploaded internal files (PDF, DOCX, TXT) without hallucinating.

Built with the PERN stack (PostgreSQL, Express, React, Node.js) and Vector Search, fully containerized with Docker.

📸 Interface Gallery

1. Landing & Authentication

The entry point to the system, featuring a modern, responsive design.

Landing Page

2. Dashboard & Workspaces

Manage your projects and see your team status at a glance.

Dashboard with created chats

3. Intelligent Chat (RAG)

Ask questions about your documents. The AI cites the context it used to generate the answer.

Chat Interface

4. Team Collaboration & Invitations

Easily expand your team. Invite colleagues to specific workspaces via email, view current members, and manage access permissions directly from the workspace settings.

Team Members & Invitations

5. Document Management

Securely upload and index corporate documents for the knowledge base.

Documents Manager

✨ Key Features

  • 🤖 RAG-Powered Chat: Uses Google Gemini Flash combined with ChromaDB to answer questions strictly based on your uploaded data.
  • 📂 Multi-Format Support: Upload and index .pdf, .docx, and .txt files automatically.
  • 🏢 Workspaces & Teams: Create isolated workspaces and invite team members via email.
  • 📱 Fully Responsive: Adaptive UI that works seamlessly on Desktop, Tablet, and Mobile.
  • 🔐 Role-Based Access: Granular permissions (Owner vs Member) for workspace management.
  • ⚡ Real-time Vectorization: Local embedding generation using @xenova/transformers (runs on CPU, no external embedding API cost).
  • 🐳 Fully Dockerized: One command to set up the Frontend, Backend, Database, and Vector DB.

🛠️ Tech Stack

Frontend

  • React 18 (Vite)
  • TailwindCSS (Custom Dark/Gold Theme)
  • Lucide React (Icons)
  • Axios (API Communication)

Backend

  • Node.js & Express
  • Multer (File Handling)
  • Pdf-parse / Mammoth (Text Extraction)

Data & AI

  • PostgreSQL 16 (Users, Chats, Metadata)
  • ChromaDB (Vector Database for Embeddings)
  • Google Gemini API (LLM for Generation)
  • Xenova Transformers (Local Embeddings)

DevOps

  • Docker & Docker Compose
  • Nginx (Reverse Proxy for Client)

🚀 Getting Started

Prerequisites

  • Docker and Docker Desktop installed.
  • A Google Gemini API Key (Free tier available).

Installation

  1. Clone the repository:

    git clone https://github.com/OkayMarta/CorpMindAI
cd CorpMindAI

  2. Configure Environment Variables: Create a .env file in the root directory:

    cp .env.example .env

    Fill in your details in .env:

    # Database Configuration
POSTGRES_USER=postgres
POSTGRES_PASSWORD=secure_password
POSTGRES_DB=corpmind_db

# Security
JWT_SECRET=random_long_string_for_security

# AI Service
GEMINI_API_KEY=your_google_gemini_key

# Email Service (for invitations)
EMAIL_USER=your_email@gmail.com
EMAIL_PASS=your_app_password

  3. Run with Docker: Start the entire application suite:

    docker-compose up --build

  4. Access the App: Open your browser and navigate to: 👉 http://localhost

🏗️ Architecture Overview

  1. Upload: User uploads a PDF.
  2. Processing: Server extracts text and splits it into "chunks" (2000 chars).
  3. Embedding: Local AI model converts chunks into numerical vectors.
  4. Storage: Vectors are stored in ChromaDB, metadata in PostgreSQL.
  5. Chatting:
    • User asks a question.
    • System converts question to vector.
    • System finds the top 15 most similar text chunks in ChromaDB.
    • System sends chunks + question to Google Gemini.
    • AI generates an answer based only on those chunks.

🛡️ License

This project is licensed under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0).

You are free to:

  • Share — copy and redistribute the material in any medium or format.
  • Adapt — remix, transform, and build upon the material.

Under the following terms:

  • NonCommercial — You may not use the material for commercial purposes (selling the app, SaaS, providing paid services using this code) without explicit permission.
  • Attribution — You must give appropriate credit.

For commercial inquiries or licensing, please contact the author.