Design and Implement Retrieval-Augmented Generation (RAG) DP-800 Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   In Azure SQL Database, which stored procedure can be used to transmit a JSON payload containing a prompt to an Azure OpenAI model?

   • Asp_send_dbmail
   • Bsp_invoke_external_rest_endpoint
   • Csp_executesql
   • Dsp_help
   Reveal Solution Discussion Select an answer first
2. 12expert · hard

   You need to send a large result set (over 100,000 rows) to a language model for summarization. The model has a token limit and cannot process the entire dataset in one request. What is the best strategy?

   • ASummarize the data in chunks using multiple calls to the model, then combine the summaries.
   • BUse FOR JSON to compress the data and send it as one JSON string.
   • CUse OPENJSON to reduce the data size before sending.
   • DSend the entire dataset in a single request and rely on the model to handle it.
   Reveal Solution Discussion Select an answer first
4. 13foundation · easy

   You receive a JSON response from a language model that contains an array of choices. You need to interpret the response by converting the JSON array into a relational format for further processing. Which Transact-SQL function is designed for this?

   • AFORMAT
   • BJSON_QUERY
   • COPENJSON
   • DJSON_MODIFY
   Reveal Solution Discussion Select an answer first
5. 14expert · hard

   A news agency stores articles in a SQL database. They want to generate summaries for each article using an Azure OpenAI model. The articles vary in length, and some are very long. The agency wants to minimize API calls while ensuring the model receives the full article content. What strategy should you use?

   • ASend the entire article in a single prompt, regardless of length.
   • BSplit long articles into chunks, summarize each chunk, and then combine the summaries.
   • CUse a fine-tuned model that has been trained on the articles.
   • DUse a model with a larger context window and send the entire article.
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A support team uses a knowledge base stored in Azure SQL Database. They want to build an internal chatbot that answers employee questions by retrieving relevant articles and sending them to a language model. The solution must minimize custom code and rely on built-in database features. What should you implement?

   • AUse sp_invoke_external_rest_endpoint to call an Azure OpenAI endpoint, passing a prompt built from retrieved articles.
   • BUse Azure Data Factory to copy the knowledge base to Azure Blob Storage and then use Azure Cognitive Search for retrieval.
   • CCreate a linked server to the language model and use OPENQUERY to send SQL queries directly.
   • DExport the entire knowledge base to a text file and upload it to Azure OpenAI Studio for fine-tuning.
   Reveal Solution Discussion Select an answer first