Prompt Engineering.
The discipline of systematically crafting prompts to get desired outputs from an LLM. Includes role-setting, few-shot examples, and constraints.
Prompt engineering is the practice of designing natural-language instructions to elicit desired behavior from an LLM. A good prompt explicitly states the role, context, format, constraints, and success criteria. The most common techniques: few-shot examples, chain-of-thought to elicit reasoning steps, role-play ("you are a..."), structured output (JSON schema), and negative instructions ("do not..."). In modern models, prompt engineering has largely become a subset of the broader practice of context engineering.
LLM'den istediğin çıktıyı almak için promptu sistematik şekilde tasarlama disiplini. Rol verme, örnek gösterme, kısıt koyma teknikleri içerir.
Prompt engineering, bir LLM'den istenen davranışı çıkarmak için doğal dilde talimat tasarlama pratiğidir. İyi bir prompt; rolü, bağlamı, formatı, kısıtları ve başarı kriterini açık şekilde verir. Tekniklerin en bilinenleri: few-shot örnek verme, chain-of-thought ile düşünme adımlarını isteme, role-play ("sen bir..."), structured output (JSON şeması), ve negative instruction ("yapma..."). Modern modellerde prompt engineering çoğu yerde context engineering'in bir alt parçasına dönüştü.
Related terms.
03Context Engineering
WorkflowDesigning what, how much, and in what order to show a model. The grown-up version of prompt engineering — the real lever in agentic systems.
Few-shot
WorkflowShowing a model a handful of examples to teach a pattern. Enough for most tasks.
Chain of Thought
WorkflowPushing a model to reason step-by-step to solve harder problems. Often as simple as "think step by step."