Coverage Matrix

This matrix maps AI-103 exam skills to the challenge set so you can target weak areas and confirm full coverage before the exam.

How to use this matrix

Use the matrix as a study checklist: start with the highest-weight domains, jump to the mapped challenges for focused practice, and finish with the capstone to validate end-to-end readiness.

Domain 1: Plan and Manage an Azure AI Solution (15-20%)

Challenges: 01-10

Skill	Challenges	Key Topics
Select the appropriate Azure AI service	01, 02	Service selection, architecture tradeoffs, Azure AI service fit
Plan and configure security (keys, RBAC, managed identity, network)	03, 04	Authentication, authorization, identity, private access
Create and manage Azure AI service resources	01, 02, 05	Provisioning, configuration, lifecycle management
Configure diagnostic logging	06	Diagnostic settings, Log Analytics, audit visibility
Manage costs	07	Pricing tiers, quotas, budgeting, optimization
Monitor Azure AI services	06, 08	Metrics, alerts, health monitoring, observability
Implement responsible AI practices	09, 10	Fairness, transparency, governance, safety
Deploy AI services in containers	05	Containers, deployment models, disconnected scenarios
Manage keys and secure endpoints	03, 04	Key rotation, endpoint protection, secret handling
Plan and implement virtual network integration	04	VNet integration, private endpoints, network isolation

Domain 2: Implement Generative AI and Agentic Solutions (35-40%)

Challenges: 11-23

Skill	Challenges	Key Topics
Create Azure AI Foundry project	11	Project setup, hubs, connections, workspace organization
Select and deploy Azure OpenAI models	12, 13	Model choice, deployments, capacity, inference options
Implement RAG (Retrieval-Augmented Generation)	14, 15	Grounding, chunking, embeddings, retrieval pipeline
Implement prompt engineering	16, 17	System prompts, few-shot design, prompt tuning
Configure content filtering	18	Safety filters, abuse monitoring, response controls
Generate code and images with Azure OpenAI	19	Code generation, image generation, multimodal use cases
Implement orchestration flows	20	Flow design, chaining, evaluation loops, process orchestration
Manage token usage and rate limits	12, 13	TPM/RPM planning, quotas, cost-aware usage
Evaluate generative AI responses	17, 20	Quality metrics, groundedness, safety, output review
Design agent architecture	21	Agent patterns, memory, planning, tool strategy
Implement tool use and function calling	22	Tool schemas, function calling, action execution
Implement multi-agent orchestration	23	Agent collaboration, delegation, workflow coordination

Domain 3: Implement Computer Vision Solutions (10-15%)

Challenges: 24-30

Skill	Challenges	Key Topics
Analyze images with Azure AI Vision	24	Image analysis, tagging, captions, detection
Implement Custom Vision image classification	25	Labels, training, evaluation, publishing
Implement Custom Vision object detection	26	Bounding boxes, detection models, scoring
Implement OCR with Azure AI Vision	27	Read API, printed text, handwriting extraction
Implement face detection and analysis	28	Face detection, attributes, analysis constraints
Analyze video with Video Indexer	29	Video insights, transcription, scene and speech analysis
Implement spatial analysis	30	People movement, occupancy, spatial event processing

Domain 4: Implement Text Analysis Solutions (15-20%)

Challenges: 31-39

Skill	Challenges	Key Topics
Analyze text (sentiment, entities, key phrases)	31, 32	NLP extraction, sentiment, entity recognition
Detect and redact PII	33	PII detection, redaction workflows, compliance handling
Translate text and documents	34	Text translation, document translation, multilingual processing
Implement speech-to-text	35	Recognition, transcription, speech ingestion
Implement text-to-speech	36	Voice synthesis, SSML, audio generation
Build CLU models	37	Intents, entities, training, deployment
Create Custom Question Answering	38	Knowledge sources, answer ranking, conversational responses
Implement speech translation	39	Real-time translation, multilingual speech pipelines

Domain 5: Implement Information Extraction Solutions (15-20%)

Challenges: 40-48

Skill	Challenges	Key Topics
Create and manage Azure AI Search indexes	40	Index schema, fields, analyzers, lifecycle
Implement scoring profiles	41	Relevance tuning, weights, boosting
Configure indexers and data sources	42	Connectors, ingestion, scheduled indexing
Implement incremental enrichment	43	Change tracking, enrichment updates, reprocessing strategy
Use built-in AI skills	44	Skillsets, enrichment pipeline, cognitive skills
Create custom skills and knowledge stores	45	Custom enrichment, projections, downstream consumption
Implement vector search and hybrid queries	46	Embeddings, vector fields, hybrid retrieval
Implement advanced queries and filters	47	Filters, facets, query syntax, result shaping
Analyze documents with Document Intelligence	48	Prebuilt models, extraction, forms and layout analysis

Capstone

Skill	Challenges	Key Topics
Integrate AI-103 skills in an end-to-end scenario	49	Solution integration, tradeoffs, production readiness review