AI & Web Glossary
Prompt engineering
Prompt engineering is the craft of writing the instructions an AI model receives so it produces accurate, consistent, on-policy results, especially when the same task runs thousands of times.
A prompt is the instruction an AI model receives. Prompt engineering is writing those instructions deliberately: what role the model plays, what information it gets, what format the answer must take, what it should refuse to do. The difference between a casual prompt and an engineered one is the difference between asking a temp to 'handle the inbox' and giving them an actual procedure.
For one-off use in a chat window, nobody needs engineering. It matters when a task runs automatically, thousands of times, unattended. A customer-facing assistant runs on a carefully built prompt that encodes your policies, your tone, and your escalation rules, and that prompt gets tested against real cases before launch the same way code does.
Practical example: an assistant that drafts replies for a services firm carries instructions like 'never quote a price; refer pricing questions to a partner' and 'always confirm the deadline mentioned in the client's email.' Those two lines prevent the two most expensive categories of mistake. That's prompt engineering: small text, large consequences.
Where this shows up in practice
- Our Custom AI Assistants & Chatbots service puts this to work for small and mid-sized businesses.
- From the blog: Building a custom AI assistant your staff will actually use