I’m working in Triple Minds and I’m working on a Candy AI Clone project like this. The project is shaping up well, but I’m stuck on a memory issue. Every time the user closes the chat and comes back, the AI forgets the previous conversation, even though I’m storing the chat in PostgreSQL.
Here’s what I’ve tried so far:
-
Saving last 20 messages as context
-
Using session IDs for each user
-
Tried caching in Redis for quick retrieval
-
Still, the AI behaves like a fresh instance after reload
I want the AI to remember past conversations while still keeping responses relevant and natural. Anyone here faced this problem while building a Candy AI or other AI clone? Tips and best practices would be super helpful