Prompt Engineering

Prompt engineering is the skill of crafting effective instructions for AI systems to generate useful, accurate responses. As educators integrate AI tools into their teaching practice, understanding these fundamentals becomes increasingly valuable.

The video “What is Prompt Engineering?” provides a concise overview of prompt engineering, explaining its significance in developing AI chatbots by crafting input prompts to achieve desired outputs.

To harness the full potential of AI in educational settings, consider these key principles when crafting your prompts:

1. Be specific and clear about your desired outcome – [Example Input]: Generate a 250-word literature review summary on culturally responsive pedagogy for a graduate-level education course syllabus.
2. Provide context relevant to educational settings – [Example Input]: I am a faculty member preparing a workshop on formative assessment strategies for new K-12 teachers. Generate a 3-point slide overview explaining key benefits.
3. Use structured formats when appropriate – [Example Input]: Create a rubric with four criteria for evaluating student reflective journals in a teacher preparation program. Use a 4-point scale and provide descriptions for each level.
4. Include examples of desired outputs – [Example Input]: Revise the following learning objective to be more measurable: ‘Students will understand classroom management theories.’ For example, ‘Students will be able to compare and contrast three classroom management theories in a written reflection.’ Now revise: ‘Students will explore inclusive teaching practices.’
5. Break complex requests into smaller steps – [Example Input]: List five recent peer-reviewed articles on teacher burnout published after 2020. Then, summarize the key findings of each article in 2-3 sentences, and note any implications for teacher education programs.

Building effective prompts requires combining several essential components that work together to guide the AI:

• Are there clear instructions for the AI to follow?
• Did you include context or background information to help frame the task?
• Are there examples demonstrating your desired outputs to show the expected format, style, or approach?
• Did you specify the format in which you want the response structured?
• Are there any constraints or limitations defined for the response?

Once you understand the fundamental elements, you can begin applying prompt engineering techniques to common educational tasks:

• Text summarization – explain in 1-2 sentences
• Information extraction
• Question answering
• Conversation – you can instruct the LLM to take on a specific behavior, intent, and identity

In this concise 8-minute video by Jeff Su, you are introduced to a structured approach for crafting effective ChatGPT prompts, focusing on six key components: Task, Context, Exemplars, Persona, Format, and Tone.

Ready to take your prompt engineering skills to the next level? Explore these more advanced techniques to further enhance your AI prompting skills and knowledge: Prompt Engineering Guide: Prompting Techniques.