Azure AI-102 glossary of model selection, Foundry, retrieval, deployment, and monitoring terms.
Use this glossary when Microsoft Certified: Azure AI Engineer Associate (AI-102) terms start to blur together. The goal is practical recognition, not encyclopedia coverage.
| Term | Exam meaning |
|---|---|
| Microsoft Foundry | The Microsoft platform surface for building, deploying, and managing AI apps, models, agents, tools, and evaluations. |
| Agent | A model-backed workflow that can use instructions, memory or context, and tools to complete tasks. |
| RAG | Retrieval augmented generation: grounding model output in external content retrieved at request time. |
| Managed identity | Azure identity used by resources to call other services without storing secrets in code. |
| Evaluation | Repeatable checks that measure quality, safety, groundedness, latency, or other behavior before release. |
| Content extraction | Turning documents, images, audio, or video into structured information for downstream use. |
| Pair | How to separate them |
|---|---|
| Foundry solution planning vs Generative AI and agents | Ask which layer the scenario is testing, then match the answer to that layer only. |
| Control vs evidence | A control changes behavior; evidence proves behavior or supports investigation. |
| Managed service vs custom build | Managed services win for lower operational effort unless the requirement needs unsupported customization. |
| Prevention vs detection | Prevention blocks or reduces a bad event; detection finds or reports that it happened. |
Do not memorize terms in isolation. For each term, write one scenario where it is the best answer, one scenario where it is a distractor, and one signal that proves it worked.