Trying to give context to AI agent

I am trying to use the AI agent in whatsapp using an api for it.

I get a request when a new message arrives, but when the next one comes, the IA does not remember the last question or answer, do you have any sugestion for this?

Example:

I say, my name is Angelo. It says: Hello angelo, how can i help you?
I ask: what is my name?
And it does not know what is it.

Also the answers are taking so long, and some times timing out.

ia 3
ia 2

It looks like your topic is missing some important information. Could you provide the following if applicable.

  • n8n version:
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app):
  • Operating system:

1.15.1 (beta)
Database i used only buffer memory and in-memory, didnt install a specific, so probably, mySQLite
Execution_process: i dont know even what is that, where do i check?
Running n8n via Docker
Operating system: idk :neutral_face:

Do you know what could it be?

It is important to mention that the data of the “Window Buffer Memory” does not persist as it is only for testing and debugging. So is it possible that you restarted n8n between the tries?
Additionally, is it important that the identical “Session Key” is set as that is how the memory identifies which context to provide.

Sadly no idea why the execution time is so long. I am however quite sure that not n8n is the reason for that but rather an external service like OpenAI. You can click on each of the nodes to see how long they did execute. You will then be able to find out which one is slowing down your execution.

1 Like

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.