The Challenge
Although IDEs like Cursor are good at generating and editing standard scripts with GenAI, it struggles with Jupyter Notebooks due to their JSON structure. This makes direct LLM-based editing easy to generate errors. As a result, educators, researchers, and developers need a specialized tool that enables precise, cell-by-cell editing of Jupyter Notebooks without sacrificing the convenience of LLM-powered suggestions.
The Jupyter Editor Solution
To address this gap, I created the Jupyter Editor—a tool dedicated to editing Jupyter Notebooks using LLM-Agent approach. Like standard text editors, it focuses on local editing rather than code execution. One can editor the Notebook here, and then run the cells in Jupyterlab.
Thanks to the LLM’s ability to use different tools—like notebook content searches, cell editing, and even web scraping—the process is both interactive and powerful. The result is a streamlined editing experience that integrates advanced AI capabilities into Jupyter Notebooks, catering to the needs of educators, researchers, and developers alike.
Demo Video
