What problem does it solve?
It solves the problem of doing multi-step Python exploration where variables and objects must persist across iterations, so you can gradually refine analysis instead of restarting from scratch each time.
Core Features & Use Cases
- Stateful Python REPL via live Jupyter kernel: keep variables, imports, and objects across executions to support true iterative workflows.
- Notebook-backed execution with editing & verification: run code inside an existing .ipynb, inspect and preview live variables, and edit cells when you need to reshape the notebook.
- Practical data-science iteration loop: ideal for checking DataFrame transformations, inspecting APIs, and incrementally building experiments like you would in a real notebook workflow.
Use case: You’re exploring a dataset and repeatedly tweak transformations (cleaning, feature engineering, aggregation), while validating intermediate variables and outputs without losing state.
Quick Start
Start a live Python kernel in a JupyterLab server and then execute code repeatedly against a scratch notebook so your variables persist across steps.
