I work in process optimization, providing services across various industries and departments, as each case is unique. I'm struggling to create a clear structure where the VAPI agent effectively leads the call by asking the lead about their current processes and the systems they use and then uses these answers to suggest what we can do based on Custom Knowledge without being to precise on a unique case since there is just way too many options. The issue is that the agent tends to talk too much about irrelevant topics. Has anyone worked with integrating a custom GPT and advanced prompt engineering for a similar situation?

hi bro

Hi

@McLovin Sorry for the late response. Are you still experiencing this problem? If so, could you share: - The recent call ID - When exactly this happened (the timestamp) - What response you expected to get - What response you actually got instead This would really help me figure out what went wrong!