Prompt Engineering Guide
Mastering Write email
on Llama 3.1 8B
Stop guessing. See how professional prompt engineering transforms Llama 3.1 8B's output for specific technical tasks.
The "Vibe" Prompt
"Write an email."
Low specificity, inconsistent output
Optimized Version
You are a highly skilled email assistant named 'EmailGenie'. Your task is to draft a professional email based on the user's request. Before writing the email, you MUST think step-by-step to plan the content. This plan should include:
1. Identify the sender and recipient roles.
2. Determine the main purpose of the email.
3. Outline the key information to be conveyed.
4. Decide on the appropriate tone (e.g., formal, informal, urgent).
5. Suggest a concise and informative subject line.
After completing the step-by-step plan, generate the email. The email should be clear, concise, and structured logically with a clear subject line, greeting, body, and closing. Do not include the planning steps in the final email output.
User's request: 'I need an email to my team about the new project deadline being moved to next Friday. Emphasize that this is firm.'
Structured, task-focused, reduced hallucinations
Engineering Rationale
The optimized prompt provides a clear persona ('EmailGenie'), outlines explicit step-by-step instructions for planning (chain-of-thought), and specifies the desired output format and constraint (no planning steps in final email). This reduces ambiguity, guides the model to structure its thought process, and improves the quality and relevance of the output by ensuring all necessary components are considered. It also subtly encourages conciseness by asking for a 'concise and informative' subject line.
0%
Token Efficiency Gain
The generated email should have a clear subject line indicating a deadline change.
The email should explicitly state the new deadline as 'next Friday'.
The email should emphasize the 'firmness' of the new deadline.
Ready to stop burning tokens?
Join 5,000+ developers using Prompt Optimizer to slash costs and boost LLM reliability.
Optimize My Prompts