jupyter-live-kernel

A stateful Python REPL exposes a live Jupyter kernel to maintain persistent variables across executions. This supports iterative data exploration and complex code building by reusing intermediate results between cells.

To start a stateful Python REPL, launch a headless JupyterLab server and connect a notebook. You can then execute Python code against the live kernel to preserve state across cells.

Yes, you need uv, JupyterLab, and a running kernel session to operate with the hamelnb tooling. These dependencies are required to establish the stateful Python execution environment.

Yes, a live Jupyter kernel is suitable for ML experimentation, API exploration, and data science workflows. The stateful environment allows step-by-step refinement and prototype data transformations across multiple cells.

A stateful Python REPL via a live Jupyter kernel maintains persistent variables and context between executions, whereas a standard environment resets state after each run. This enables iterative coding without redefining variables or reloading data.

Your Python code loses context because standard execution environments are stateless. Using a stateful Python REPL with a live Jupyter kernel preserves variables across executions, allowing you to prototype data transformations and reuse intermediate results.