Term
Prompt engineering
Prompt engineering is the discipline of steering AI models through carefully crafted input prompts — closer to precise writing than to classic programming.
Prompt engineering — explained in more detail
Prompt engineering is the practice of steering a language model’s behaviour by shaping its input. Unlike classic programming, no deterministic code is written; instead the goal is to communicate role, task, format and examples clearly enough that the model is most likely to produce a usable answer.
Established techniques
Few-shot prompting hands the model several examples of the desired behaviour directly in the prompt. Chain-of-thought prompts the model to make intermediate steps visible before answering. Role definitions (“You are an experienced editor …”) set a stylistic frame. Structured output specifications — JSON schemas, fixed section headings — make results machine-readable.
Distinction from similar terms
Prompt engineering is not programming in the classic sense: less logic, more precise writing with system. It is broader than the system prompt: prompt engineering covers the user prompt and examples too — the entire input context, not just the fixed behaviour instruction.
Entdecke mehr
Effort level and deep thinking: two independent axes for AI tasks
Effort scales breadth, deep thinking scales depth. When each setting makes sense — with three clear examples and one rule of thumb.
GlossarChain-of-Thought
Chain-of-Thought (CoT) is a prompting technique that asks the model to spell out its reasoning in intermediate steps — boosting accuracy on multi-step tasks.
LexikonPrompt Security — Injection, Leaking, Guardrails
How prompt injection, prompt leaking, and jailbreaks work — and which defenses (guardrails, spotlighting, sanitization) actually help.