Hi @Vapi team, we're noticing after making many prompts updates to refine the Assistant it starts to get glitchy... is this normal? We are thinking we should duplicate it now but then we're worried if we do that then it goes back to square one of how it functioned at the start which was very unpredictable for the first couple hundred calls, then sort of ironed out a bit. If we were to duplicate the Assistant, do their learnings start over again? Will they need time to smooth out, learn the KB again etc? Or will their learnings and improvements be copied into the new duplicate?

Hey! To help track down this issue, could you share: - The call ID - What response you expected to get - What response you actually got instead This would really help us figure out what went wrong!

Hi Shubman, This issue does not have an exact call ID - it is a general question we were hoping to gain clarity on. Any insight on this?

@firefly_0206 The assistants are LLM based so they don't learn from past conversations unless your update their prompt or pass the past conversations history. So, your pretty much safe on this. if anything comes up just ping me,