Getting Started with CDEs
Your complete quickstart guide to cloud development environments - from zero to productive in minutes
Before You Start
Any Computer
CDEs work from any machine with a browser or SSH client - even a Chromebook or tablet.
Internet Connection
A stable internet connection is required. Most CDEs work well on connections 10Mbps+.
Git Repository
Your code hosted on GitHub, GitLab, Bitbucket, or any Git provider.
New to Cloud Development?
No worries - you don't need to change how you code. CDEs feel like local development but run in the cloud. Your favorite IDE, keybindings, and extensions all work the same way.
5-Minute Quickstart (GitHub Codespaces)
GitHub Codespaces is the fastest way to try CDEs. As of July 2026, the GitHub Free plan includes 120 core-hours and 15 GB-month of storage per month; GitHub Pro includes 180 core-hours and 20 GB-month.
Core-hours are not wall-clock hours
A core-hour is one CPU core running for one hour, so the machine size you pick divides your allowance. 120 core-hours is 60 hours on a 2-core machine, or 30 hours on a 4-core machine. Plenty of write-ups get this wrong, and teams that budget core-hours as calendar hours run out early.
Beyond the included allowance, compute is billed per hour by machine size - 2-core $0.18, 4-core $0.36, 8-core $0.72, 16-core $1.44, 32-core $2.88 - plus $0.07 per GB-month of storage. Rates as of July 2026; check the GitHub Codespaces billing documentation for current figures.
Go to Your Repository
Navigate to any GitHub repository you want to work on.
Press the Period Key
On any GitHub repo page, press . (period) to open github.dev instantly.
Or click the green "Code" button and select "Codespaces" tab.
Create a Codespace
Click "Create codespace on main" - your environment will be ready in ~30 seconds.
Start Coding
You're now in a full VS Code environment. Run commands in the terminal, install extensions, and commit changes.
Choose Your Platform
Different platforms suit different needs. Here's how to get started with each major CDE provider.
GitHub Codespaces
Best for GitHub-centric teams wanting zero-config cloud development.
- 120 core-hours/month free on GitHub Free
- Deep GitHub integration
- No infrastructure needed
Ona (formerly Gitpod)
AI-powered development platform with managed environments and autonomous coding agents.
- Core plan from $20/month; Enterprise is custom-priced
- Built-in AI coding agents
- Enterprise-only runners in your own AWS or GCP VPC
There is no customer-managed self-hosted Ona. Enterprise runners keep source code and execution in your network, but Ona operates the control plane - bring-your-own-cloud, not self-hosting.
Get StartedCoder
Genuinely self-hosted platform with Terraform templates and AI agent integration.
- Community edition: free, open source, self-hosted
- Any cloud or on-prem
- Premium tier is annual per-user, priced on request
DevPod
Open-source desktop app for running DevContainers anywhere, from vCluster Labs (renamed from Loft Labs in August 2025).
- Free and open source under MPL-2.0
- Works with any provider
- Desktop + CLI
Okteto
Kubernetes-native development environments you can self-host or run in your own cloud account.
- Open source and genuinely self-hostable
- Starter plan free up to 5 seats
- Bring-your-own-cloud deployment
VS Code Remote SSH
Connect VS Code to any Linux machine via SSH - DIY approach.
- Free (your infra costs)
- Full VS Code experience
- Maximum flexibility
Names you may still see recommended, but should not start with
- Daytona - pivoted away from cloud development environments and now sells sandbox infrastructure for running AI-generated code. Its public repository became unmaintained in June 2026 when core development moved to a private codebase. Still a reasonable agent sandbox; not a CDE.
- AWS Cloud9 - closed to new customers on July 25, 2024 and in maintenance mode. No end-of-life date has been announced, so existing users are not stranded, but it is not something to adopt now.
- Amazon CodeCatalyst - closed to new customers on November 7, 2025.
- JetBrains Space and JetBrains Fleet - both discontinued. Space cloud was deactivated on June 1, 2025 and Fleet downloads ended on December 22, 2025. JetBrains Gateway, covered above, is unaffected.
Connect Your Favorite IDE
VS Code
Most popular choice
Install the "Remote - SSH" extension from the marketplace
Press F1 and select "Remote-SSH: Connect to Host"
Enter your workspace SSH connection string
Host my-workspace
HostName workspace.example.com
User developer
Browser-Based (No Install)
Access from any device
Most CDEs include a web-based VS Code (code-server) that runs directly in your browser. Just click a link and start coding - no installation required.
Setting Up DevContainers
DevContainers define your development environment as code. Add one to your repo and everyone gets the same setup.
Create .devcontainer/devcontainer.json
{
"name": "My Project",
"image": "mcr.microsoft.com/devcontainers/base:ubuntu",
// Install Node.js and Python
"features": {
"ghcr.io/devcontainers/features/node:1": {},
"ghcr.io/devcontainers/features/python:1": {}
},
// VS Code extensions to install
"customizations": {
"vscode": {
"extensions": [
"esbenp.prettier-vscode",
"dbaeumer.vscode-eslint"
]
}
},
// Commands to run after container starts
"postCreateCommand": "npm install",
// Ports to forward
"forwardPorts": [3000, 8080]
}Set Up AI Coding Assistants
AI coding assistants have become standard equipment on most teams. CDEs let you centrally manage, govern, and pre-install them for everyone.
GitHub Copilot
Inline code suggestions and chat integrated into VS Code, JetBrains, and Codespaces.
$10/mo for Pro. Free for verified students and open source maintainers.
Cursor
AI-native editor built on VS Code. Connects to CDE workspaces via Remote SSH for full cloud development.
Free tier available. Pro at $20/mo with unlimited AI completions.
Claude Code
Agentic CLI tool that runs directly in your CDE terminal. Understands full codebases and makes multi-file changes.
Requires Anthropic API key or Max subscription. Ideal for CDE terminals.
Why AI Assistants Work Better in CDEs
CDEs provide consistent, powerful compute for AI tools. Your team gets uniform AI access, API keys are managed centrally through secrets, and AI-generated code stays within your security boundary. Pre-install assistants via DevContainers so every workspace is AI-ready from the start.
Common First Tasks
Set Up Git Credentials
Most CDEs inherit credentials from your SSO login.
Forward Ports
CDEs auto-detect and forward ports to your browser.
Install Extensions
Your extensions install on the remote machine, not locally. Add them to devcontainer.json to share with your team.
Configure Secrets
Never put secrets in devcontainer.json. Use your CDE's secret management:
- - Codespaces: Repository secrets
- - Coder: Workspace parameters
- - Ona: Environment variables
Common Issues & Solutions
Connection drops or is slow
- - Check your internet connection stability
- - Try a wired connection instead of WiFi
- - Reduce the workspace machine type if not needed
- - Contact your IT team about firewall settings for port 22 (SSH)
Extensions not working
- - Extensions run remotely, not locally - reinstall them in the workspace
- - Some extensions require specific dependencies - check extension docs
- - UI extensions need to be installed locally, workspace extensions remotely
Git authentication issues
- - Use SSH keys instead of HTTPS for Git operations
- - Enable credential forwarding in your CDE settings
- - For Codespaces: GitHub credentials are automatic
- - For others: Add your SSH key to the workspace or use SSH agent forwarding
Workspace won't start
- - Check if you've exceeded your quota (Codespaces: 120 core-hours/month on GitHub Free, 180 on Pro)
- - Try rebuilding the container (may fix corrupted state)
- - Check for syntax errors in devcontainer.json
- - Review the creation logs for specific error messages
Next Steps
Master DevContainers
Create advanced configurations, multi-container setups, and team templates.
Learn moreAI Coding Assistants
Set up Copilot, Cursor, Claude Code, and other AI tools in your CDEs.
Explore AI toolsCompare Platforms
Deep-dive comparison of Coder, Ona, Codespaces, and more.
Compare toolsPlan Your Migration
4-phase roadmap for rolling out CDEs to your entire organization.
View roadmap