Glossary term
Glossary term
Infrastructure and Serving
A network-addressable location (typically a URL) where a service can be reached.
Created for this library
An ML platform team exposes one inference endpoint per production model so application teams integrate via a stable HTTP contract.
A SaaS company versions its model endpoints and routes a fraction of traffic to the new version before full cutover.
An enterprise ML platform documents its endpoints in an internal portal so product teams can self-serve integration without filing tickets.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License