Prompt engineering is the discipline of crafting targeted inputs for large language models so they consistently produce high-quality outputs. Unlike classical programming, natural language is the medium – complemented by techniques such as role instructions, clear examples, step-by-step reasoning ("chain-of-thought") or structured output formats.
In practice, prompt engineering often makes the difference between a useful tool and an unreliable toy. Good prompts provide context (who is asking, in what domain), task (what exactly should happen), format (how the answer should look) and examples.
In recruiting, prompt engineering is used for job evaluations, CV summaries, cover-letter drafts or interview questions. Filtering inappropriate outputs – for example involving sensitive personal characteristics – is also shaped through carefully designed prompts.
Lunigi uses carefully curated prompts to summarise roles understandably, deliver AI-safety assessments and match profiles against real requirements.