Forked from@ashavega/My notes from fixing a customer-support copilot
My notes from fixing a customer-support copilot
Not magic wording, just enough context and a clear definition of done. Easy to adapt.
I am stuck on a customer-support copilot. I need to draft a grounded support reply using policy text and flag anything that needs a human, but my current attempt is getting vague and too big. What would you check first?
First, shrink the decision. Do not solve the whole subject at once. 1. Put the source of truth in the prompt. 2. Mark missing evidence instead of filling it in. 3. Add a stop condition before tool use. Write down the result you need by the end of this session. Then choose one input you trust and make a first pass that can be checked. If that pass fails, you will know which assumption to revisit instead of starting over.