Built this: LabX
A Docker image that runs JupyterLab and a browser-based VS Code (code-server) side by side on one shared Python environment and workspace.
On 10 August 2026 I made the first commit of LabX, a self-contained Python lab environment in Docker.
What you get
From the README:
- JupyterLab 4.6.2 at
localhost:5959 - LabX-Server, a branded code-server 4.132.0 (VS Code in the browser), at
localhost:5960 - Python 3.11.9 on Debian Bookworm slim
- a preinstalled stack: NumPy, Pandas, scikit-learn, Keras, Transformers, Accelerate, OpenCV, NLTK, spaCy, MLflow, W&B, Stable-Baselines3, Gymnasium and more
- 17 VS Code extensions bundled as
.vsixfiles and installed offline - password protection for both services, configurable by environment variable
- multi-architecture builds (
linux/amd64,linux/arm64) and a mounted, persistent/workspace
Where it connects
It’s the container version of the environment I set up by hand at the start of this blog: a venv, Jupyter and an editor, packaged so it can be rebuilt anywhere.
Sources
- SahajIVVIX-1/LabX (GitHub) — GitHub
- First commit 8fef865 — GitHub
This is a summary written from the linked public sources. Figures and numbers are as reported there.