Skip to main content
VS Code provides a polished editing experience on top of GRID sessions. Once the containers are running, you can attach directly to grid_core, browse notebooks, and develop Python code against the same filesystem that Jupyter exposes. This page covers two scenarios:
  • You are on a managed or remote machine and already connected via VS Code Remote-SSH.
  • You are running GRID on your own workstation and have VS Code installed locally.
The workflow is identical either way—just make sure the grid_core container is running (init … in the CLI) before attaching.

​
Prerequisites

  • VS Code with the Remote Development extension pack (Remote-SSH and Dev Containers).
  • GRID containers running on the target machine (grid_core, grid_server, simulator).
  • Optional: an existing remote VS Code window (for managed or remote nodes) as described in the Managed GRID guide.

​
1. Connect to the Host (If Remote)

If you are working on a remote machine:
  1. Use Remote-SSH to connect (see Managed GRID Enterprise).
  2. Wait for the remote VS Code window (green status bar).
If you are working locally, just open VS Code on the same machine where GRID is installed.

​
2. Attach to the grid_core Container

  1. Ensure the containers are up (init airgen or init isaac from the CLI).
  2. In VS Code, open the Command Palette (F1 / ⌘⇧P / Ctrl+Shift+P).
  3. Run Dev Containers: Attach to Running Container….
  4. Pick grid_core from the list.
  5. VS Code opens a new window attached to that container. The root folder defaults to /workspace, which contains:
    • sample_notebooks/ — same content exposed via Jupyter
    • Any volumes you mounted (for example /workspace/datasets)
    • Logs and artifacts generated by your sessions
The first attach installs the VS Code server inside the container, which takes a few seconds.

​
3. Explore the Workspace

  • Use the Explorer to open notebooks (*.ipynb) or Python scripts.
  • Launch terminals (Terminal > New Terminal) to run Python, CLI tools, or Git commands directly inside the container environment.
  • VS Code inherits all dependencies baked into grid_core, so packages, CUDA binaries, and environment variables match what your sessions use.

​
4. Develop and Iterate

  • Edit notebooks or scripts in VS Code and run them using the built-in Jupyter support or the integrated terminal.
  • Use the Python extension for linting, breakpoint debugging, and refactoring.
  • Changes to mounted volumes (for example /workspace/datasets) are reflected across containers, Jupyter, and your local machine transparently.

​
5. Detach Safely

When you finish:
  • Close the VS Code window attached to grid_core. The container keeps running until you terminate it via the CLI.
  • If you are on a remote VM, you can keep the Remote-SSH window open for later or disconnect via the status bar menu.
With the container attachment workflow, VS Code becomes a first-class IDE for GRID—perfect for longer editing sessions, pair programming, or integrating Git workflows. Combine it with the Hello GRID Enterprise tutorial for a hands-on introduction, then keep iterating directly inside the container.