📚 Glossary

Prompt Engineering

Prompt engineering is the practice of crafting and optimizing text inputs to language models to consistently produce higher-quality, more accurate, or better-formatted outputs.

Definition¶

Prompt engineering is the discipline of designing effective inputs (prompts) to language models to reliably produce desired outputs. A well-engineered prompt specifies the task, provides relevant context, constrains the output format, and may include examples — all within the model's context window.

Why Prompt Quality Matters¶

The same underlying model can produce dramatically different outputs depending on how a request is framed. "Write a summary" produces a different result than "Write a 3-sentence executive summary of the following text, using plain English appropriate for a non-technical audience."

Prompt engineering is particularly valuable for: - Consistent output formatting (always produce a table, always use bullet points) - Style and tone control (match the voice of a specific document) - Complex multi-step reasoning tasks - Reducing hallucination by asking the model to stick to provided context

Key Techniques¶

Role prompting — assigning the model a persona: "You are an expert copy editor who specializes in making AI-generated text sound natural."

Few-shot prompting — providing examples of input-output pairs before the actual task:

Input: [AI text sample] → Output: [humanized version] Input: [AI text sample] → Output: [humanized version] Now do this: [your text]

Chain-of-thought — asking the model to reason step-by-step before giving a final answer, which improves accuracy on logical tasks.

Output constraints — specifying exactly what format the output should take, its length, and what to exclude.