Deploy CollabAI on Your Server – A Self-Hosting Guide

CollabAI is a powerful AI-enhanced collaboration tool designed to bring knowledge management, team productivity, and automation under one roof. Whether you want to build internal knowledge bases, create custom GPT-powered agents, or manage intelligent workflows, CollabAI has you covered.

This guide walks you through self-hosting CollabAI — from prerequisites to complete setup — so you can take full control of your data and environment.

---

🧠 What is CollabAI?

CollabAI is a two-part application:

• Frontend (Client) – React-based UI for users and admins.

• Backend (Server) – Node.js API handling authentication, AI interactions, database ops, and more.

It uses MongoDB for primary data storage and Pinecone for storing and querying vector embeddings for AI-based search and knowledge retrieval.

---

🌟 Use Cases

Here are just a few ways teams can benefit from CollabAI:

• 🔍 AI-Powered Knowledge Base: Upload documents, websites, or FAQs and use GPT to answer questions instantly.

• 💬 Smart Agent Assistants: Train AI agents with specific knowledge and deploy them within teams.

• 🛠 Internal Automation: Automate workflows like ticket responses or customer inquiries.

• 📊 Data-Driven Insights: Use vector similarity and semantic search to surface the right content at the right time.

---

✅ Prerequisites

Before starting, make sure you have:

• ✅ A Unix-based server (Ubuntu/Debian recommended)

• ✅ MongoDB (local or Atlas)

• ✅ A Pinecone account with:

  • API Key

  • Index for storing embeddings

• ✅ Node.js v20+ installed

• ✅ npm or yarn installed globally

---

📦 Step-by-Step Deployment

1. Clone the Repository

git clone https://github.com/sjinnovation/CollaborativeAI.git

Navigate to the project directory 

cd CollaborativeAI

---

2. Set Up the Frontend

Navigate to the frontend directory 

cd client

Run the below command to install node

curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.2/install.sh | bash

Export the path

export NVM_DIR=”$([ -z “${XDG_CONFIG_HOME-}” ] && printf %s “${HOME}/.nvm” || printf %s “${XDG_CONFIG_HOME}/nvm”)” [ -s “$NVM_DIR/nvm.sh” ] && \. “$NVM_DIR/nvm.sh” # This loads nvm

Install the node version 

nvm install 20

Install the frontend dependencies

npm install

Create and Configure the Environment File:

cp .env.sample .env vim .env

Update the .env file with the backend URL and port, for example:

VITE_BACKEND_URL=http://localhost:8011

---

3. Set Up the Backend

cd ../server npm install

Prepare Environment Configuration:

cp .env.sample .env.local vim .env.local

Here’s an example .env.local configuration:

JWT_SECRET=dummyJWTSecret123 MONGO_URI=mongodb+srv://<username>:<password>@<cluster-url>/ PORT=8011 CLIENT_URL=http://localhost:4000 NODE_MAILER_USER=myemail@example.com NODE_MAILER_PASS=myemailpassword ADMIN_EMAIL=admin@example.com MAILGUN_API_KEY=dummyMailgunAPIKey123 MAILGUN_DOMAIN=example.com CC_EMAILS=cc1@example.com,cc2@example.com MONGODB_NAME=mydatabase AWS_ACCESS_KEY_ID=dummyAWSAccessKey123 AWS_SECRET_ACCESS_KEY=dummyAWSSecretKey123 AWS_REGION=us-west-2 PINECONE_API_KEY=your-pinecone-api-key PINECONE_ENVIRONMENT=your-pinecone-environment PINECONE_INDEX_NAME=your-pinecone-index

💡 Create your vector index from the Pinecone Console

---

4. Run Both Services Using PM2

Make sure PM2 is installed globally:

npm install -g pm2

Start Backend

cd server pm2 start npm –name “collab-ai-backend” –namespace collab-ai — start — -p 8011

Start Frontend

cd ../client pm2 start npm –name “collab-ai-frontend” –namespace collab-ai — start — -p 4000

Check running services:

pm2 status

---

5. Initialize the Application

Create a Super Admin user via cURL:

curl -X POST http://localhost:8011/api/init \   -H “Content-Type: application/json” \   -d ‘{   “fname”: “Super”,   “lname”: “Admin”,   “email”: “superadmin@example.com”,   “password”: “Bolbona”,   “employeeCount”: 100,   “companyName”: “INIT_COMPANY” }’

Let’s Explore with the http://localhost:4000 with the above created super admin email and password. 

---

🔒 Security Tips

• Use strong secrets in your .env.local file.

• Never expose your .env.local in version control.

• Set up HTTPS (e.g., using NGINX + Let’s Encrypt).

• Regularly back up your MongoDB and Pinecone index data.

---

📚 Conclusion

CollabAI is a flexible and powerful tool to supercharge internal knowledge and AI capabilities within your organization. With the steps above, you’ll have a production-ready instance of CollabAI hosted on your own infrastructure — giving you full control and endless customizations.

---

📬 Need Help?

Have questions or want to contribute? Open issues or submit pull requests via the GitHub repo.