AI engineering glossary · Prompt context engineering
What is structured extraction?
Structured extraction (also called Information extraction with LLMs): Using a model to pull specific fields out of unstructured text, like names, dates and amounts, and return them in a fixed structure. It replaces brittle rules for documents that vary widely in layout.
Structured extraction: an example
From a WhatsApp order message, extract {"item": "kurta", "size": "M", "pincode": "560001"}.
Related terms
More Prompt context engineering terms
- Structured output
- System prompt
- Task decomposition
- Template variable
- Thinking budget
- Token budget
- Token counting
- Token pricing
Where you learn it
Structured extraction is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.