Databricks Certified Generative AI Engineer Associate — Question 79

A Generative AI Engineer received the following business requirements for an internal chatbot. The internal chatbot needs to know what types of questions the user asks and route them to appropriate models to answer the questions. For example, the user might ask about historical failure rates of a specific electrical part. Another user might ask about how to troubleshoot a piece of electrical equipment.

Available data sources include a database of electrical equipment PDF manuals and also a table with information on when an electrical part experiences failure.

Which workflow supports such a chatbot?

Answer options

• A. Parse the electrical equipment PDF manuals into a table of question and response pairs. That way, the same chatbot can query tables easily to answer questions about both historical failure rates and equipment troubleshooting.
• B. The chatbot should be implemented as a multi-step LLM workflow. First, identify the type of question asked, then route the question to the appropriate model. If it’s a historical failure rate question, send the query to a text-to-SQL model. If it’s a troubleshooting question, then send the query to another model that summarizes the equipment-specific document and generates the response.
• C. There should be two different chatbots handling different types of user queries.
• D. The table with electrical part failures should be converted into a text document first. That way, the same chatbot can use the same document retrieval process to generate answers regardless of question types.

Correct answer: B

Explanation

Option B is correct as it outlines a structured approach where the chatbot first identifies the question type and then routes it to the appropriate model, ensuring accurate responses based on the inquiry. Option A suggests combining data into a single format, which may not effectively differentiate the types of questions. Option C proposes separate chatbots, which complicates the system unnecessarily. Option D assumes a uniform document retrieval process, which may not be suitable for varied question types.