Heads up! To view this whole video, sign in with your Courses account or enroll in your free 7-day trial. Sign In Enroll

The 5 Principles of Effective AI Prompting - AI Foundations Day 3

Well done!

You have completed The 5 Principles of Effective AI Prompting - AI Foundations Day 3!

Sign up for Treehouse Back to Library

Principle 3: Provide Examples

2:42 with Travis Alstrand

Sometimes showing beats telling. The escalation ladder from zero-shot to few-shot, the four situations where examples genuinely help, how many to use, and why example quality matters far more than quantity with modern models.

Vocabulary

  • XML tags — Angle-bracket labels like and used to fence off sections of a prompt. Especially effective with Claude, and the recommended way to mark anything untrusted.
  • Zero-shot / few-shot prompting — Zero-shot is asking with no examples — always try this first. Few-shot means adding one to three examples of good output to show the pattern you want.

Related Discussions

Have questions about this video? Start a discussion with the community and Treehouse staff.

Sign up

    Related Discussions

    Have questions about this video? Start a discussion with the community and Treehouse staff.

    Sign up

    Giving the model 1-3 examples of what good output looks like is called few-shot 0:00
    prompting. 0:03
    The Escalation Ladder. How to Approach Any Task 0:06
    Step 1. 0:09
    Go zero-shot first. 0:10
    Just describe what you want. 0:12
    No examples. 0:14
    Try the simplest thing possible. 0:15
    Step 2. 0:17
    Go few-shot if you need to. 0:19
    Add 1 to 3 input-output examples if zero-shot isn't cutting it. 0:20
    Step 3. 0:26
    Only add more complexity when the output still isn't good enough. 0:27

    You need to sign up for Treehouse in order to download course files.

    Sign up

      You need to sign up for Treehouse in order to set up Workspace

      Sign up