Good AI Prompting

17 Feb, 2026

Notes from the Google IT Security course on Coursera, which included a module titled, "Streamline Workflows with AI."

Prompting is the input you provide the AI model to generate a desired response. A strong AI prompt follows the T-C-R-E-I framework, which is:

• Task
• Context
• References
• Evaluate
• Iterate

You can remember these AI prompting steps through the mnemonic phrase: Thoughtfully Create Really Excellent Inputs.

Task

• What do you want the model to do?
• You can break the task into two: Persona & Format.
• Persona: What expertise do you want the AI to draw from?
• Format: How do you want the output to appear?

Context

• The necessary details to help the AI understand what you need from it.

References

• Add references for the AI to use while creating it's output.

Evaluate

• Did the input give you the output you needed?

Iterate

• If you didn't get what you needed, try again by adding more information or editing your prompt.

AI is best used by applying the human-in-the-middle approach. This approach combines machine learning with human intelligence to train, use, verify, and refine AI results. The human-in-the-middle approach means being mindful of what you put into the AI tool, and always evaluating and verifying what comes out.