Craft Effective Prompts GH-300 Practice Questions (Page 1)

Questions 1–5

1. 1application · medium

   A team is using GitHub Copilot Chat to refactor a legacy codebase. They want Copilot to suggest a new implementation for a specific module, but they are concerned that Copilot might not understand the existing architecture. What should they include in their prompt to ensure the suggestions are relevant?

   • AThe full source code of the entire codebase.
   • BA brief description of the module's purpose and the key constraints of the existing architecture.
   • CA request to ignore the existing architecture and propose a completely new design.
   • DA list of all the files in the repository without any explanation.
   Reveal Solution Discussion Select an answer first
2. 2application · medium

   A developer is using GitHub Copilot to generate a function that converts a string to title case. The developer wants the output to follow a specific pattern: each word capitalized, but with a few exceptions like 'and' and 'of'. Which prompt approach is most effective?

   • AProvide a few examples of input/output pairs showing the desired transformation, including the exceptions.
   • BAsk the model to use its pre-trained knowledge of title case rules.
   • CDescribe the desired behavior in a single sentence without examples.
   • DProvide a single example and rely on the model to generalize the pattern.
   Reveal Solution Discussion Select an answer first
4. 3application · medium

   A developer is using GitHub Copilot to generate a SQL query. The first attempt returns a query that is close but not exactly what they need. What is the best next step?

   • ARefine the prompt by adding more specific requirements and constraints.
   • BAccept the query and manually adjust it later.
   • CAsk the same prompt again, hoping for a different result.
   • DSwitch to a different coding language.
   Reveal Solution Discussion Select an answer first
5. 4foundation · easy

   Why is iterating on prompts considered a best practice?

   • AIt allows the model to learn from its mistakes.
   • BIt increases the model's pre-trained knowledge.
   • CIt helps refine the prompt based on the model's responses to improve quality.
   • DIt reduces the need for context in future prompts.
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   In a GitHub Copilot Chat session, a developer has been discussing a specific project. They now ask a question that is unrelated to the previous conversation. How should they handle this to get the best response?

   • AStart a new chat session to avoid the previous context influencing the response.
   • BProvide a brief summary of the previous conversation to give context.
   • CUse a system instruction to clear the conversation history.
   • DAsk the question in the same session, assuming Copilot will ignore the previous context.
   Reveal Solution Discussion Select an answer first