If I deploy a Streamlit app behind an nginx reverse proxy, will it scale to many users or is the backend blocking (or threaded but inefficient) when there are many requests?

I am not sure if it is rather a tool thought for local usage to have simple UIs for your data science / AI projects or if it can be deployed at a larger scale. I am not talking about scaling to multiple machines and more, but it should not have wait times for parallel requests.