Quickstart Guide

Get up and running with Cirron CLI in just a few minutes. This guide will walk you through installing the CLI, creating your first project, and deploying it.

Prerequisites

Node.js (v20.19 or higher) - for the CLI
Python (3.11 or higher) - for the project

Installation

Install the Cirron CLI globally:

npm install -g @cirron/cli

Verify the installation:

cirron --version

A curl one-line installer and a Homebrew formula are coming. For now, npm is the supported distribution channel.

Step 1: Create Your First Project

Create a new ML project with an interactive setup:

cirron init

The CLI will guide you through:

Project name
Framework selection (PyTorch, TensorFlow, scikit-learn)
Model type (Classification, Regression, etc.)
Sample data inclusion
Jupyter notebook setup

Or create a project with specific options:

cirron init my-first-model --template pytorch --git

Note: You can use the CLI without authentication for local development. Authentication is only needed to connect your project to the Cirron platform.

Step 2: Navigate to Your Project

cd my-first-model

You can run the generated project directly with python train.py and then python serve.py, or drive it through the CLI as shown below.

Step 3: Test Your Project

Run the test suite to ensure everything is working:

cirron test

This will test:

Environment compatibility
Requirements validation
Unit tests
Model loading
Data validation

Step 4: Build Your Project

Build your project for deployment:

cirron build

This creates a production-ready build of your ML model.

What You’ve Built

Congratulations! You’ve successfully: ✅ Installed Cirron CLI
✅ Created a new ML project
✅ Tested your project
✅ Built your project
✅ Connected to Cirron platform (optional)
✅ Deployed your project (optional)

Next Steps

Explore Your Project

Your project structure includes: For a PyTorch project:

my-first-model/
├── cirron.yaml           # Project configuration
├── train.py              # Training entry point
├── serve.py              # Serving entry point
├── requirements.txt      # Python dependencies
├── models/               # Trained model files
├── artifacts/            # Build and run artifacts
└── build/                # Build output

The exact files depend on the template you chose.

Development Workflow

Make changes to your model or code
Test locally with cirron test
Build with cirron build
Deploy with cirron deploy (coming soon)

Advanced Features

Local development: Work entirely offline with full CLI functionality
Platform integration: Connect to Cirron for deployment and monitoring
Multiple environments: Deploy to staging and production
Environment variables: Manage configuration per environment
Rollback: Quickly revert to previous deployments
Monitoring: View logs and project status

Troubleshooting

Installation Issues

# Check Node.js version
node --version

# Reinstall CLI
npm uninstall -g @cirron/cli
npm install -g @cirron/cli

Project Issues

# Check project configuration
cat cirron.yaml

# Reinitialize project
cirron init --force

Multiple models in one repo?

If you keep several models in a single repository, add a root cirron.yaml with a workspace: key and the CLI can discover and validate all of them from the repo root. See Monorepo support.

Need Help?

Documentation: Check the CLI documentation
Commands: Explore all available commands
Support: Contact support at support@cirron.com

Ready to build your next ML project? Check out the full CLI documentation for advanced features and workflows.

​Quickstart Guide

​Prerequisites

​Installation

​Step 1: Create Your First Project

​Step 2: Navigate to Your Project

​Step 3: Test Your Project

​Step 4: Build Your Project

​What You’ve Built

​Next Steps

​Explore Your Project

​Development Workflow

​Advanced Features

​Troubleshooting

​Installation Issues

​Project Issues

​Multiple models in one repo?

​Need Help?