Quick Note — Choose the appropriate Foundry services for generative tasks, grounding, vector search, agent workflows, or multimodal processing
AI-103 › Unit 1: Plan and manage an Azure AI solution › Choose the appropriate Foundry services for generative AI and agents › Choose the appropriate Foundry services for generative tasks, grounding, vector search, agent workflows, or multimodal processing
Quick Note — Choose the appropriate Foundry services for generative tasks, grounding, vector search, agent workflows, or multimodal processing
An expense-approval build needs three things: fields out of scanned receipts, a screen for jailbreak attempts, and answers grounded in procurement policy. A team that reaches for one frontier model and a long prompt gets all three approximately. The same build assembled from Document Intelligence, Content Safety, and AI Search gets typed fields with a maintained schema, a purpose-built injection detector, and hybrid retrieval with a semantic ranker — three guarantees instead of three hopes.
Retrieval target
- Choose the appropriate Foundry services for generative tasks, grounding, vector search, agent workflows, or multimodal processing
- Closed-book recall
- 5 minutes
- Open Microsoft Learn after a miss
Decision anchors
| Prompt | Compact answer |
|---|---|
| Grounding and vector search | Azure AI Search — full-text, vector, hybrid, and multimodal queries; integrated vectorization generates embeddings in-pipeline; semantic ranker reorders by meaning. It also underpins Foundry IQ, the managed knowledge layer that turns enterprise content into reusable, permission-aware knowledge bases. |
| Structured extraction from documents | Azure Document Intelligence in Foundry Tools — prebuilt models (prebuilt-invoice, prebuilt-receipt, prebuilt-layout, prebuilt-read) with schemas defined and maintained by Microsoft, returning strongly typed values. |
| Any modality into a shape you define | Azure Content Understanding — processes "unstructured data of any type (image, documents, audio, video)" and extracts structured insights against pre-defined or user-defined formats. |
| Safety screening | Azure AI Content Safety — four harm categories, Prompt Shields for direct and indirect attacks, protected material detection, custom blocklists. |
| Agent workflows | Foundry Agent Service — the runtime that "manages conversations, tool calls, and agent lifecycle". Choose it when something must be decided at runtime: pick a tool, act, read the result, decide again. One prompt and one answer does not need it. |
| Resource type matters | A Foundry resource provides unified access to models, agents, and tools through a project endpoint. An Azure OpenAI resource provides only the /openai/v1 endpoint — no agents, no platform tools. Choosing the wrong resource type makes tools unreachable by configuration, not by setting. |
Read the answers once, then cover the right-hand column and reconstruct each one from the prompt. A useful answer names the requirement, the capability that satisfies it, and the nearest alternative it rejects. If you can only recognize the answer after seeing it, retrieval is not yet secure.
Ninety-second explanation
Without notes, explain:
- Which service provides hybrid retrieval, and what does the semantic ranker add on top?
- What distinguishes Content Understanding from Document Intelligence?
- Which requirement in a stem tells you an agent runtime is needed rather than a model call?
- Why can a build on an Azure OpenAI resource fail to attach file search or code interpreter?
- What is Foundry IQ, and what does permission-aware buy you?
Then check yourself against current Microsoft Learn. Record the missing decision rule, not merely the missed product name, in your error log.
Choose the appropriate Foundry services for generative tasks, grounding, vector search, agent workflows, or multimodal processing — quick retrieval
When to go deeper
Go deeper when the choice depends on a service's current preview status or regional availability — agentic retrieval carries region restrictions that classic search does not. Start from Introduction to Azure AI Search and the Foundry Tools overviews.
Source and freshness
Written against current Microsoft Learn documentation for the AI-103 skills measured (16 April 2026). Reviewed 2026-08-20. Microsoft Learn controls changing product contracts — availability, limits, preview status, naming, and retirement dates move, and this note is deliberately compact.