Remote Development
Five practical ways to develop remotely on a VKS cluster — SSH, browser-based VS Code, Jupyter Notebook, VS Code Dev Containers, and DevPod
LLM-era workloads need too much GPU / RAM / disk to develop comfortably on a laptop. The "edit locally, run remotely" pattern is taking over. The five tutorials below cover the common remote-dev paths, ordered by entry barrier.
VS Code over SSH
Classic flow — local VS Code + Remote-SSH extension into the container
VS Code in the browser (code-server)
Nothing to install locally — just a browser
Jupyter Notebook remote
Default for data science and model experiments
VS Code Dev Containers
Connect to Pods via the Kubernetes extension — no SSH required
DevPod (containerized dev environments)
Standardize dev environments with devcontainer.json
Last updated on
Was this page helpful?