mljar/mercury

Mercury is a framework that transforms Jupyter notebooks into interactive web applications without requiring callbacks, frontend code, or notebook rewrites.

The problem Mercury solves is the gap between notebook-based data analysis and shareable web applications. Developers and data scientists often create analyses in Jupyter but struggle to share them as polished, interactive web apps. Mercury bridges this by letting you add simple Python decorators and components to your existing notebook code. The framework handles the conversion to a web interface automatically, preserving the notebook's logic while making it accessible through a browser without exposing the underlying code or requiring users to understand Python.

Mercury suits data scientists and analysts who want to share interactive analyses with non-technical stakeholders without learning web development. It works well for dashboards, data visualizations, and exploratory tools where the core logic already exists in a notebook. The tool is particularly valuable when you need to publish work quickly without restructuring your analysis code. Teams should consider Mercury when they want to avoid the overhead of building a separate web application while maintaining the flexibility of notebook-based development.

The project shows consistent development activity with regular updates addressing user feedback and expanding capabilities. The maintainers actively respond to issues and incorporate feature requests from the community. Development focuses on improving the ease of converting notebooks to web apps and expanding the range of interactive components available. The project maintains clear documentation and examples that help users get started quickly. There is ongoing work to enhance the visual appearance and customization options of generated applications.