Introduction

Software engineers often need to understand complex codebases quickly. AI can assist in this process if prompted correctly.

This lesson will guide you through crafting prompts that help AI provide meaningful insights into code.

Understanding Code with AI

AI can be a powerful ally in code comprehension, offering explanations, summaries, and even identifying potential issues. The key is to ask the right questions.

Components of an Effective Code-Related Prompt

• Specificity: Be precise about the code segment you need help with. You can directly reference the code segment by inserting the snippet after your initial prompt.

• Goals: Clearly state what you want to understand or achieve. Examples include; ‘I want to debug this code.’, ‘Please help me fix the following error: [Error here]. This is my code: [Code here]’

• Context: Provide necessary background information about the code’s functionality. For example, if you are debugging a script, providing the entire script as context will be the most useful.

Example Prompt:

"I’m getting the following error in Python: IndexError: list index out of range. I’m trying to loop through a list and print its elements, but the error happens when I try to access the list index. Here's my code:

 for i in range(5):

    print(my_list[I])

Could you help me fix this and explain what’s causing the issue?"

This prompt is specific about the problem, includes the relevant code snippet, and clearly states the goal (fixing the error). It also provides context about what you're trying to do with the list.

Crafting Your Prompt

When crafting a prompt for code understanding, consider the following steps:

1. Isolate the Code Segment: Focus on a specific part of the code to avoid overwhelming the AI.

2. Describe the Purpose: Explain what the code is intended to do.

3. Ask Targeted Questions: Formulate questions that lead to a deeper understanding of the code’s logic and structure.

Let's put this into practice, with debugging

Debugging is an essential skill for software engineers, and AI can be a valuable asset in this process. This lesson focuses on how to formulate effective queries that help AI assist you in debugging tasks.

AI can help identify bugs, suggest fixes, and even automate some debugging tasks. To do this effectively, it needs to be guided by clear and precise queries.

Components of an Effective Debugging Query

• Description of the Issue: Clearly describe the bug or issue you’re encountering.

• Code Context: Provide relevant snippets of code where the issue occurs.

• Expected vs. Actual Results: Explain what the code should do and what it’s actually doing.

• Steps Taken: Detail any steps you’ve already taken to try to resolve the issue.

Writing Your Debugging Query

When designing a debugging query for AI, consider the following steps:

1. Isolate the Problem: Narrow down the issue to the smallest reproducible case.

2. Provide Relevant Information: Include error messages, code snippets, and system details.

3. Ask for Specific Help: Whether it’s understanding an error or finding a bug, be clear about what you need.

Activity

Choose a coding problem you’ve encountered or create a hypothetical bug.

Formulate a query to the AI detailing the issue and asking for assistance. Evaluate the AI’s response and the effectiveness of your query.

OR

Choose a piece of code you’re unfamiliar with.

Craft a prompt asking the AI to explain the code’s functionality, suggest improvements, or identify bugs. Analyze the AI’s response for clarity and accuracy.