Data Flow Canvas – Privacy-first visual data science platform

Summary: Data Flow Canvas is a browser-based data science platform that runs all processing locally using WebAssembly, ensuring data privacy. It offers 257 drag-and-drop blocks for data transformation, machine learning, statistics, and visualization, with full Python support including pandas, NumPy, and scikit-learn. Users can collaborate in real time with hidden IP addresses and export pipelines as standalone Python code.

What it does

It enables data analysis and machine learning entirely in the browser without server uploads, using WebAssembly to run Python libraries locally. Users build workflows by combining drag-and-drop blocks covering various data science tasks.

Who it's for

Data scientists, analysts, and spreadsheet users working with sensitive or private data who require local processing and collaboration.

Why it matters

It solves privacy concerns by eliminating the need to upload data to external servers while providing a full-featured, open source data science environment.