Generative AI: Prompt Engineering Tips for Working with LLMs
When you interact with a Large Language Model (LLM) like ChatGPT and ask a question, have you ever wondered why the response might not meet your expectations?
The answer could lie in how you craft your prompts. Here are some best practices for prompt engineering:
1. Ask the AI to role-play (e.g., "Act as my assistant")
2. Be specific (e.g., "Write a 100-word email")
3. Provide context (e.g., "This email is an invitation to my child's birthday party and will be sent to the parents of my child’s friends.")
4. Request detailed output (e.g., "The email should start with 'Dear ABC' and end with 'Sincerely XYZ.' It should include three paragraphs: introductory, body, and closing")
5. Ask for citations (where applicable)
6. Iterate and refine
What has your experience been like when prompting an LLM? I would love to hear your thoughts. Please share them in the comments.
Generative AI: Prompt Engineering Tips for Working with LLMs
When you interact with a Large Language Model (LLM) like ChatGPT and ask a question, have you ever wondered why the response might not meet your expectations?
The answer could lie in how you craft your prompts. Here are some best practices for prompt engineering:
1. Ask the AI to role-play (e.g., "Act as my assistant")
2. Be specific (e.g., "Write a 100-word email")
3. Provide context (e.g., "This email is an invitation to my child's birthday party and will be sent to the parents of my child’s friends.")
4. Request detailed output (e.g., "The email should start with 'Dear ABC' and end with 'Sincerely XYZ.' It should include three paragraphs: introductory, body, and closing")
5. Ask for citations (where applicable)
6. Iterate and refine
What has your experience been like when prompting an LLM? I would love to hear your thoughts. Please share them in the comments.