QUESTION #1: Why do we want to support streaming? What about streaming is important, or useful? Streaming is important for voice applications where TTFT (time to first token) would be needed otherwise there will noticeable pauses in the voice responses. QUESTION #2: Why are we using User Session here? What about Python makes us need to use this? Why not just store everything in a global variable? Sessions are useful when there are more than one user. Keeping everything in one global variable will surely cause issues as many users will be updating the global variable as they have shared data/state. It will be confusing when the application runs.