Question

A retail bank wants to build a customer-facing conversational agent that answers questions strictly from its internal library of PDF policy documents and product brochures. The team wants the agent to cite information from these documents rather than rely only on the foundation model's general knowledge. Which Vertex AI Agent Builder capability should they configure to meet this requirement?

Accepted Answer

Connect a data store that indexes the bank's documents so the agent grounds its answers in that content

Answer

Increase the model temperature so the agent generates more detailed policy explanations

Answer

Fine-tune a foundation model in Model Garden on unrelated open-source financial datasets

Answer

Expand the model's context window so it can memorize all documents permanently