The practice of designing and optimizing input prompts to reliably elicit desired outputs from an LLM. Effective prompts combine clear instructions, relevant context, output format specification, and examples.
Prompt engineering is the art of communicating effectively with LLMs. Core techniques include: Chain-of-Thought (ask the model to think step by step), Few-Shot (provide 2-5 examples), System Prompts (set role and constraints), and Structured Formatting (XML tags, Markdown headers).
Anthropic's Claude responds particularly well to XML tags in system prompts (<instructions>, <context>, <output_format>) and explicit role definitions. The clearer the output format specification, the more consistent the results.
Anthropic's AI assistant with industry-leading reasoning and safety
OpenAI's AI assistant powering 100M+ users worldwide
Weekly AI tool reviews, news digests, and how-to guides.
Join 12,000+ builders