AI Engineering Glossary
Use this glossary when a lesson, paper, model card, or code review introduces a term faster than it explains it. Search by the exact term or an alias, read the direct definition first, then use the practical note to connect it to a system you can build.
Each entry belongs to one learning category. Related terms gives you the next useful concepts without forcing a fixed path. Definitions describe the common engineering meaning, but provider-specific behavior can differ. When an API contract or model card disagrees with a general definition, the current official documentation wins.
The twelve categories are: Math & training; Models & inference; Data & representations; Retrieval & generation; Prompting & context; Agents & tools; Evaluation & safety; AI-native development; Infrastructure & serving; Reliability & operations; Security & governance; Multimodal systems.