Our data team kept hitting the same problem. We needed UIs for business users, dashboards, CRUD apps, internal tools. Streamlit was our go to but it reruns the entire script on every interaction, the interfaces all look identical, and it gets painful fast when real users are involved.
React was too much overhead for a data team.
So we built burq, a Python UI compiler that compiles to pure Vanilla JS + HTML. You write Python, run burq build, and get static files. FastAPI serves them. Deploy to Databricks Apps.
What changed for us: frontend work went from days to hours. Business users stopped complaining. And everything deploys cleanly on Databricks Apps with no extra runtime overhead.
It is open source and on PyPI. Would love feedback from others building internal tools on Databricks.
Would love to hear if you build something with it or need help building with it
GitHub: github.com/itsdaniyalm/burq
Docs: burq.dev
