AI engineering glossary · Statistics inference
What is guardrail metric?
Guardrail metric: A metric monitored during an experiment to ensure a change does not damage something important, even if the main target metric improves. A guardrail breach can block a launch. Teams typically set an acceptable threshold for each one before the test starts.
Guardrail metric: an example
A test that boosts sign-ups also tracks page-load time and refund rate as guardrails.
Related terms
More Statistics inference terms
- Heteroscedasticity
- Homoscedasticity
- Instrumental variable
- Interaction term
- Jackknife
- Kolmogorov-Smirnov test
- Likelihood function
- Likelihood ratio test
Where you learn it
Guardrail metric is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.