Prompt Engineering (Beginner Level)

← Back to “AI for Good” Nonprofit Workshop Resources

Chiv Heng Consulting — www.chivheng.consulting • LinkedIn

</aside>

Last reviewed: July 2026

Effective use of generative AI is not about focusing on any specific tool. You may find through experimentation that you have a preference for some tools over others. The most critical aspect is developing your skills in prompting the AI tool and reviewing and editing what it produces for quality and accuracy.

Work with AI chat bots as if you are communicating with a smart, eager intern on their first day:

Your intern generally has "high intelligence and low wisdom."
They have read the manual, but they don't know your specific context.
If you give them vague instructions, you will get generic work.
If you manage them with clear context, specific goals, and feedback, they will be the most productive assistant you've ever had.

Core Principles of Prompt Engineering

1. Role: Tell the AI who it is (e.g., "Act as a Grant Writer").

2. Context: Explain why you need the information.

3. Output: Describe the format, required content, and quality bar. Say what good looks like, such as a concise board-ready update with clear takeaways (Table, Bullet points, "something I can copy & paste with formatting into Google Docs").

4. Constraints: Set limits (e.g., "Max 200 words," "No jargon," "Use US Currency formatting") to keep the AI focused.

5. Focus: When working in an AI chat session, keep that session focused on a singular task, otherwise you risk confusing the AI with too much context that may not be directly relevant to the specific task you've asked it to do. (e.g., don't ask AI to help with Budget Review and Program Evaluation in the same chat session)

6. Definition of done: Write one sentence describing what a finished, usable answer looks like before you prompt, and check the result against it.

Worked example: a board update

Use this prompt with synthetic program notes: