AI Best Practices

In this course, AI tools (such as ChatGPT, Copilot, Gemini, or Claude) can be powerful aids for brainstorming, debugging, and scaffolding. However, you are responsible for understanding the material, verifying correctness, and ensuring your work meets course standards.

AI Reflection (Required for Each Assignment — 5 Points Total)

For each assignment, you must complete an AI Reflection worth 5 points:

4 points for the written reflection in README.md
1 point for submitting your AI interaction log (CSE3300_AI_LOG.md)

The reflection must be at least 2–3 sentences per section (more if needed). It should briefly but clearly explain how you used AI before, during, and after your project.

Answer the following questions in your README.md file.

Before Coding

What is the goal of this assignment?
When will you use AI, and when will you avoid it?
What conceptual questions did you ask the AI?

During Development

Paste your three most useful AI prompts.
What was the AI’s response? (Summarize.)
What did you change in the AI’s output, and why?
What worked and what did not? (Be specific.)

After Completion

What errors did the AI make that you caught?
What debugging or testing did you do?
What did you understand better because of the AI?
What would you change about how you use AI next time?

This reflection demonstrates that you used AI as a learning tool, not as a replacement for doing the work yourself.

Note: Before writing any code, complete the Before Coding section. This reflection is required for every assignment.

AI Interaction Log (Required — 1 Point)

In addition to the reflection, you must submit an AI interaction log.

Your CSE_3300_AI_LOG.md file should include copy-pasted prompts and responses from any AI tools you used (e.g., ChatGPT, VS Code Copilot, Copilot Agents). This file is graded for completion only and is used to verify AI usage.

Verification and Responsibility

When using AI, you are still fully responsible for your submission. This means you must:

Verify correctness

Test AI-generated code or explanations in your own environment.
Fix any errors or gaps in logic yourself.

Check for security issues

AI may suggest unsafe practices.
You must review and correct these.

Maintain style and readability

Clean up AI-generated code to match course expectations (proper formatting, comments, descriptive names).

Warning: Failure to review AI outputs is the same as submitting incorrect work.

Branch Working Code (Strongly Recommended)

When you begin the AI-assisted portion of your assignment, create a new Git branch so your original working code is preserved. This prevents losing progress if issues arise with AI-generated changes and makes it easier to compare before-and-after results.

AI Best Practices

Contents