/

Prompt engineering

GLOSSARY

AI & automation

Prompt engineering

Prompt engineering

Prompt engineering

THE SHORT VERSION

Prompt engineering is the process of designing and testing instructions and context to guide an AI model’s response.

Prompt engineering is the process of designing and testing instructions and context to guide an AI model’s response.

Prompt engineering involves crafting and refining the instructions and context given to a generative AI model. It can include specifying the task, providing examples and stating the expected response format. The prompt is tested and revised by reviewing the output it produces. Source: IBM: Prompt engineering

Write a brief that can be checked

For a destination summary, provide the approved source material, audience, length and Travelgenix writing rules. Ask the model to identify information it cannot support. These instructions give the reviewer clear criteria for checking the draft.

Keep useful test cases

Save the prompt alongside several test inputs, including one with missing or conflicting details. Repeat those cases after editing the prompt. Comparing the results makes it easier to see whether an improvement for one example creates a problem elsewhere.

Written and updated by Travelgenix. Sources checked on 19 October 2026. Practical advice and examples are Travelgenix editorial guidance.

Sources

The official pages behind this explanation. This is general information, not legal advice, and rules change, so check the latest version.