Glossary
Evaluation
Evaluation is testing how well an AI system does a task by checking its answers against a set of cases with known good results.
What it means in practice
Build your own small test set from real tasks and rerun it whenever you change a prompt or a tool. A tool that looks good in a demo can still fail on your own work.
Example
Ten real customer questions with the answers you accept, used to compare two chatbots.
Guide: Test chatbots on your own 10 tasks before you payTerm: BenchmarkAll terms