A regional bank wants a chat assistant that answers account and product questions. It should follow the bank's policy wording and draw on its published help articles. The team wants to configure this through a console and plain-language guidance, not by writing an intent classifier. Which Google Cloud offering is most appropriate?
Show answer and explanation
Answer: B
Agent Builder provides a managed way to create conversational agents by describing their behavior in natural language and connecting them to content such as documents or websites. The grounding keeps responses tied to the bank's material. Model Garden is a catalog of models, Pipelines orchestrates ML workflows, and Endpoints serve a model that the team would still have to build.
Why the other options are wrong
- A. Model Garden is a catalog of foundation and open models. It supplies model choices but not a configured conversational agent with grounding.
- C. Pipelines automate ML workflow steps such as training and evaluation. They do not create a customer-facing chat assistant.
- D. Hosting a self-built intent classifier means building the language understanding yourself. The requirement is to avoid coding and to ground answers in documents.