This repository is the main backend for the startup template. For reference implementations, check out our other templates:

• DataClouder Angular Template
• DataClouder Node Template

Description 📝

A NestJS-based backend API template designed to power AI learning and conversation applications. This project serves as the backend infrastructure for an interactive platform that demonstrates advanced AI conversation features and Ionic mobile development.

🎯 Key Features:

• AI conversation management
• Google Cloud Services integration
• Scalable architecture for learning applications
• Ready-to-use authentication system

💡 Note: While the project is tightly integrated with Google Cloud services, these modules can be easily removed if not needed.

Getting Started 🚀

Clone the Project

git clone https://github.com/dataclouder-dev/startup-template-node [your-project-name]

or use the button on github right top corner CREATE TEMPLATE

✅ Prerequisites

• node >= 22.0.0
• Make >= 3.0.0 (Optional but highly recommended)
• Docker (Optional)
• Env File, rename .env.example to .env and add your environment variables
• Google Cloud credentials and environment variables- Service account credentials go to:
  • If you created a firebase project you already have the Google Cloud, got to the link https://console.cloud.google.com/iam-admin/serviceaccounts?project=[your-project-id] and create a service account
  • Give Editor Permissions (Unless you know what you are doing then give minimal privileges) and download the json file
  • Copy in .cred/keys.json
• MongoDB credentials
  • Get the mongo variables, will be easy if you extract from conection string and copy in .env
  • example mongodb+srv://admin:MyPass1234@cluster0.5q5kdhg.mongodb.net/my_db_name
  • MONGO_USER=admin
  • MONGO_PASS=MyPass1234
  • MONGO_HOST=cluster0.5q5kdhg.mongodb.net
  • MONGO_DB=my_db_name

Initial Setup

1. Clone the repository
2. Configure your environment:
   • Copy .env.example to .env
   • Add your environment variables
   • Create .cred folder and add Google Cloud credentials
3. Customize the user model for your needs

Development 💻

Install Dependencies

Run the Project

# Development mode
npm run start

# Watch mode (recommended for development)
npm run start:dev

# Production mode
npm run start:prod

Testing 🧪

# Unit tests
npm run test

# End-to-end tests
npm run test:e2d

# Test coverage
npm run test:cov

Customization 🛠️

User and Firebase

You can handle users easialy, create google credentials, for Gcp service, firebase included, and save in .cred/keys.json, check .env.example for variables you may need.

GOOGLE_APPLICATION_CREDENTIALS=./.cred/keys.json

Module Management

You can remove unnecessary modules by:

1. Deleting the module folder
2. Removing the import from app.module.ts

Deployment 🌐

Local Docker Testing

Before deploying, test locally with Docker:

docker build -t dc-template-node .
docker run -p 8080:8080 --env-file ./.env dc-template-node

Docker Compose

For a quick setup using the pre-built image from GHCR:

This will start the container using ghcr.io/adamofig/dataclouder-dev-node:latest.

• Port: Access the app at http://localhost:7991 (configurable via HOST_PORT in .env).
• Configuration: Ensure you have your .env file and .cred/key-qa.json ready.

Cloud Deployment

Manual Deployment Steps

1. Authenticate with Google Cloud:

2. Submit the build: i recommend use the make commands i created to automate, but you can use manual gcloud commands if you have the knowledge.

• First set environments in makefile or .env, you can leave IMAGE_NAME, REGION, SERVICE_NAME, but PROJECT_ID is uniqui to your project

  make gcp-enable-services # will ask to enable bill on GCP this is required to deploy
  make build-push #
  make deploy-service

3. Next time you can just run:

Automated Deployment With Cloud Build

Note: before try to automate the deployment, i highly recommend do one manual deployment to check if everything is working. specially becouse for every cloud run service, variables need to be set first time, consicutive times no need.

1. Fork the repository
2. Go to cloud build and create a new trigger
3. Grant github access, select the repository and accept conditions
4. Add seetings for the trigger to your needs
5. Optional: Add permissions to the service account, Logs Writer, Cloud Run Admin or log only default logs
6. Add the repository in artifact registry (recommended add policies to remove old versions)

Example Deployment Command

gcloud builds submit --tag gcr.io/ai-learning-dev/node-app-image .
gcloud run deploy ai-learning-service \
  --image gcr.io/ai-learning-dev/node-app-image \
  --platform managed \
  --region us-central1 \
  --allow-unauthenticated

Clone and Reuse template for new projects

git clone https://github.com/adamofig/dataclouder-template-node [new-project-name]
cd [new-project-name]
npm install

Unistall what you don't need

npm uninstall @dataclouder/conversation-card-nestjs @dataclouder/lessons-nestjs @dataclouder/storage-uploader

remove module from app module. you are ready to go.

Contributing 🤝

Contributions are welcome! Please feel free to submit a Pull Request.

License 📄

MIT License