I am a python developer and coordinator at Siemens Energy. I have been following streamlit for a long while now, and figured I would write some feedback based on my experience.
Where streamlit shines:
The marketing of streamlit focusses strongly on data science and this makes sense, but I would argue it really shines in replacing corporate excel sheets. There is a way bigger market for this imo. My primary use for streamlit has been to replace complex multi-user excel sheets with nice and tidy streamlit apps connected to a database. This saves tremendous amounts of time by automating large parts of data entry and improves quality.
Really the only features I feel streamlit still lacks here are interactive tables with triggers (like PYQT) and tabs.
Deployment:
I think streamlit would be worth money to us and it has already proven its worth. Sadly no corporation like Siemens will ever allow cloud servers. Fortunately I found out I could deploy via docker and nginx.
I love that streamlit allows this. I would never get any kind of reoccurring monthly fee approved, so I urge you to stay away from this.
Business model:
The open source factor and local deployment factor are crucial for us to use streamlit.