AI-102: Azure AI engineering, historical course

Objectives and progression

Six modules with fictional APS and project examples, explained practice, and an internal assessment of 24 decisions in 60 minutes. Distinguishes classic and current Foundry contracts. AI-102 and the Azure AI Engineer certification retired on 2026-06-30. Historical syllabus dated 2025-12-23; this course retains technical learning with current notes and is not complete AI-103 preparation. Historical format states 100 minutes and score 700/1000, not equivalent to 70%, with no fixed question count inferred.

Audience: Application engineers, APS teams, and technical managers of AI solutions.

Prerequisites: Azure and AI fundamentals, HTTP APIs, identity, and development basics.

300 estimated study minutes

• Prepare access, capacity, and diagnosis for an AI production solution.
• Build traceable answers and evaluate complete application behavior.
• Retain authorization and operational accountability for agent-proposed actions.
• Choose vision tasks and acceptance criteria around the errors that matter.
• Preserve meaning and apply fallback before turning language into actions.
• Trace data from source to answer and validate each pipeline boundary.

Modules

1. Services, identity, and operations
2. Generation, grounding, and evaluation
3. Agents, tools, and control
4. Vision, metrics, and lifecycle
5. Language, speech, and integration
6. Search, extraction, and evidence

Continue learning

Data / AI

• AI-900: Azure AI Fundamentals (historical)
• AZ-900 — Azure Fundamentals
• DP-900: Azure Data Fundamentals

References and version

AI-102 historical skills measured 2025-12-23; exam retired 2026-06-30

• AI-102 historical objectives and retirement · 2026-09-30
• Retired Azure AI Engineer credential historical format · 2026-09-30
• AI-103 distinct current syllabus · 2026-09-30
• Foundry Tools authentication · 2026-09-30
• Private network and DNS access · 2026-09-30
• Container connectivity billing and disconnected options · 2026-09-30
• Classic Azure OpenAI Actions and DataActions · 2026-09-30
• Classic Azure OpenAI quota rate limits and retries · 2026-09-30
• Classic service monitoring metrics · 2026-09-30
• Classic content filtering behavior · 2026-09-30
• Classic model fine tuning · 2026-09-30
• Generative solution evaluation and observability · 2026-09-30
• Classic application tracing and sensitive recording · 2026-09-30
• Classic prompt flow inputs and development · 2026-09-30
• Classic deployment options · 2026-09-30
• Agent function request application execution and output · 2026-09-30
• Retry limits idempotency and uncertain outcomes · 2026-09-30
• Custom Vision model quality and representative data · 2026-09-30
• Custom Vision published prediction iteration · 2026-09-30
• Custom Vision retirement 2028-09-25 · 2026-09-30
• Video Indexer extracted insights · 2026-09-30
• PII detection and redacted output · 2026-09-30
• Dynamic dictionary scope and language requirements · 2026-09-30
• SSML speech synthesis controls · 2026-09-30
• Speech recognition error evaluation · 2026-09-30
• CLU intents entities and retirement 2029-03-31 · 2026-09-30
• CLU evaluation and None intent · 2026-09-30
• Question answering confidence and fallback · 2026-09-30
• Indexer source enrichment and output mappings · 2026-09-30
• Blob deletion detection and index synchronization · 2026-09-30
• Skillset enrichment graph and projections · 2026-09-30
• Hybrid keyword and vector retrieval · 2026-09-30
• Semantic reranking of retrieved candidates · 2026-09-30
• Search security filtering by user access · 2026-09-30
• Document Intelligence composed models · 2026-09-30
• Asynchronous document analysis and result polling · 2026-09-30
• Content Understanding multimodal structured extraction · 2026-09-30
• Technical exam scaled scoring · 2026-09-30
• Retrieval augmented generation · 2026-09-30
• Model catalog capability and availability · 2026-09-30
• Vision capabilities and Image Analysis 4.0 lifecycle · 2026-09-30
• Direct and indirect prompt attacks · 2026-09-30
• Extraction confidence and evaluation · 2026-09-30
• Tokens embeddings and transformer attention · 2026-09-30