I’m wondering what the suggested way is to handle a multi-user chat/task session with Smolagents.
The agent has its own memory, so for a single user, I can control it by sending the proper ‘reset_memory’ value. But how do I handle this with multiple concurrent users?
I guess I can return the agent’s memory, update it, and send it back, but this seems cumbersome. There must be a simpler way.
Any suggestions are welcome