Prompt Engineering

Back to Large Language Models

Crafting inputs to get desired outputs from LLMs without modifying model weights. A critical skill for working with foundation models.

Key Techniques

  • Zero-shot — task description only, no examples
  • Few-shot — provide examples in the prompt
  • Chain-of-Thought — instruct model to reason step by step
  • System Prompts — set behavior, persona, constraints

nlp llm prompt-engineering