2023-03-06
Notion AI blog post as an intoduction to prompt engineering.
Written by: Joseph Reilly
Prompt engineering is the skill of designing prompts or questions that elicit a desired response or outcome. In this blog post, I will discuss three key aspects of prompt engineering: framing the prompt, providing the right level of constraint, and avoiding biases.
How you frame a prompt has a significant impact on the responses you will get. Carefully consider the words and phrases you use to ensure your prompt elicits the type of response you are looking for. Subtle changes in wording can steer responses in very different directions. Frame prompts as open-ended questions to encourage creative and thoughtful answers.
Too much constraint can limit responses, but not enough constraint can lead to unfocused answers. Find the right balance for your goals. Constrain prompts by specifying a length, format, audience, purpose, or other parameter. However, leave enough flexibility for personal expression and a diversity of options.
Unconscious biases can easily creep into prompts. Review prompts to ensure they are neutral, inclusive, and do not lead respondents in a particular direction. Avoid prompts that make assumptions about gender, race, ability, or other personal attributes. Carefully consider the images, examples, and stereotypes used in prompts. Biased prompts will skew the responses you receive.
Prompt engineering is a crucial skill for eliciting meaningful responses and achieving your goals. By thoughtfully framing prompts, providing the right level of constraint, and avoiding biases, you can craft prompts that lead to actionable insights and quality results. With practice, prompt engineering will become second nature.