ScriptsApr 16, 2026·3 min read

DevPod — Open-Source Dev Environments on Any Infrastructure

DevPod by Loft Labs creates reproducible, containerized development environments that run on any cloud, Kubernetes cluster, or local Docker — all from a single devcontainer.json spec.

TL;DR
DevPod spins up reproducible dev environments from devcontainer.json on Docker, Kubernetes, AWS, GCP, Azure, or SSH targets.
§01

What it is

DevPod by Loft Labs is an open-source tool that creates reproducible development environments based on the devcontainer.json standard. Unlike proprietary cloud IDEs, DevPod is client-only and infrastructure-agnostic. You pick the provider (Docker, Kubernetes, AWS, GCP, Azure, SSH) and DevPod handles provisioning, connecting your IDE, and cleanup.

It targets developers who want GitHub Codespaces-like convenience without vendor lock-in. Platform teams use DevPod to standardize development environments across their organization while letting developers choose their own infrastructure.

§02

How it saves time or tokens

DevPod eliminates the 'works on my machine' problem by ensuring every developer runs the same container environment regardless of their local OS or cloud provider. Onboarding a new team member goes from a day of setup to a single devpod up command.

Because environments are disposable and reproducible, developers can spin up isolated workspaces for each feature branch and tear them down when done, avoiding dependency conflicts between projects.

§03

How to use

  1. Install the DevPod CLI:
curl -L -o devpod "https://github.com/loft-sh/devpod/releases/latest/download/devpod-linux-amd64"
chmod +x devpod && sudo mv devpod /usr/local/bin/
  1. Create a workspace from any Git repository:
devpod up github.com/my-org/my-repo
  1. Connect to the running workspace:
devpod ssh my-repo
§04

Example

A devcontainer.json that DevPod uses to build the environment:

{
  'name': 'Node.js Dev',
  'image': 'mcr.microsoft.com/devcontainers/javascript-node:20',
  'forwardPorts': [3000],
  'postCreateCommand': 'npm install',
  'customizations': {
    'vscode': {
      'extensions': ['dbaeumer.vscode-eslint']
    }
  }
}

DevPod reads this spec and provisions a container with Node.js 20, installs dependencies, and forwards port 3000 to your local machine.

§05

Related on TokRepo

  • DevOps Tools -- Infrastructure automation tools that complement container-based workflows
  • Self-Hosted Tools -- Tools for running development infrastructure on your own servers
§06

Common pitfalls

  • Provider configuration is required before first use. Run devpod provider add docker (or your preferred provider) before devpod up or the command fails with no provider error.
  • Large devcontainer images increase startup time significantly. Use slim base images and layer caching to keep workspace creation under 2 minutes.
  • SSH key forwarding must be configured for private repository access inside the workspace. Use devpod up --ssh-agent-forwarding to forward your local SSH agent.

Frequently Asked Questions

How does DevPod compare to GitHub Codespaces?+

GitHub Codespaces is a managed service tied to GitHub. DevPod is client-only and infrastructure-agnostic -- you run environments on Docker, Kubernetes, or any cloud provider. Both use the devcontainer.json standard, so configurations are portable between them.

Which IDEs does DevPod support?+

DevPod supports VS Code (local and remote), JetBrains IDEs (via Gateway), and any terminal-based editor through SSH. The desktop app provides a GUI for managing workspaces across all supported IDEs.

Can DevPod run on Kubernetes?+

Yes. DevPod has a Kubernetes provider that creates workspace pods on your cluster. This lets platform teams offer development environments on shared infrastructure with resource limits and namespace isolation.

Is DevPod free?+

Yes. DevPod is fully open-source under the Apache 2.0 license. There is no paid tier for the CLI or desktop app. Loft Labs offers a commercial platform (Loft) for enterprise features like access control and cost management.

Does DevPod support GPU workloads?+

Yes, when using providers that support GPU passthrough (Docker with NVIDIA runtime, Kubernetes with GPU nodes, or cloud instances with GPUs). You configure GPU access in the devcontainer.json or provider settings.

Citations (3)
  • DevPod GitHub— DevPod creates reproducible dev environments based on devcontainer.json
  • Dev Containers Spec— Dev Containers specification for reproducible development environments
  • DevPod Documentation— Loft Labs maintains DevPod as open-source under Apache 2.0

Discussion

Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.

Related Assets