PyGWalker bets notebooks want Tableau-style drag-and-drop; 16k stars

Kanaries is a data-tools company that built Graphic Walker, an open-source alternative to Tableau, then packaged it as PyGWalker ('Python binding of Graphic Walker'): one pyg.walk(df) call turns a pandas dataframe in a Jupyter notebook into an interactive drag-and-drop visual-analysis UI with cleaning, filtering, chart building and natural-language querying. The project launched on Hacker News on 2023-02-20 with 712 points and 61 comments.

The bet was that data scientists would rather explore data with a Tableau-style interface inside the notebook they already use than switch between scripting charts and a separate BI tool. The library stayed Apache-2.0 open source while Kanaries built the surrounding business: a 0.6 release line with reusable Walker objects, code export and DuckDB-backed local computation, support for Colab, Kaggle, Databricks, Streamlit, marimo and Hex, plus paid products such as PyGWalker Cloud and runcell, an AI code agent for Jupyter. A June 2024 arXiv paper documented the approach.

By the 2026-09-05 fetch of the repository, PyGWalker showed 16.0k stars, 887 forks and 774 commits, with cloud computation and GPT-powered features marketed to the same notebook audience. The open-source library is the distribution wedge; the company is selling the data-AI layer around it.