AI-200: Developing AI Cloud Solutions on Azure
Browse certifications
Interactive Study Guide
Each module covers one exam topic with plain-English explanations, real-world scenarios, and built-in practice. Everything you need to understand and retain the material — no tab-switching required.
Domain 1: Develop containerized solutions on Azure ›
Domain 2: Develop AI solutions by using Azure data management services ›
Domain 3: Connect to and consume Azure services ›
Domain 4: Secure, monitor, and troubleshoot Azure solutions ›
Exam Resources
Official learning paths, exam details, skills measured, and community resources to supplement your study.
About the AI-200 Exam
Build the back end of AI on Azure: containers, vector data, messaging, and observability.
AI-200 is the new associate-level Microsoft developer certification — focused on the back-end of an AI solution. Where AI-102 / AI-103 test you on the AI itself (model selection, agents, RAG design), AI-200 tests you on wiring AI into a production application — containers (ACR, Container Apps, AKS, KEDA), AI data services (Cosmos DB NoSQL, PostgreSQL + pgvector, Azure Managed Redis), messaging (Service Bus, Event Grid, Functions), and security + observability (Key Vault, App Configuration, OpenTelemetry, KQL).
⚠️ AI-200 replaces AZ-204 (retiring 31 July 2026). If you’re starting Azure developer prep today, study AI-200 — your AZ-204 path will run out of runway.
Who Should Take This Exam?
The AI-200 is designed for Azure developers who already write back-end code and want to formalise the AI-integration skills the industry now expects. The audience profile is explicit: contributing to all phases of implementing AI solutions, with an emphasis on back-end services and components.
You should be comfortable with:
- Azure SDKs and third-party SDKs used in Azure
- Azure data management services (Cosmos DB, PostgreSQL)
- Azure monitoring and troubleshooting (Log Analytics, Application Insights)
- Azure messaging and eventing (Service Bus, Event Grid, Event Hubs)
- Vector databases (pgvector, Cosmos vectors, RediSearch)
- Python programming
- Implementing containerised applications on Azure
Typical study time: 4-8 weeks of part-time study, depending on Azure background.
Exam Quick Facts
| Detail | Value |
|---|---|
| Exam Code | AI-200 |
| Title | Developing AI Cloud Solutions on Azure |
| Level | Associate |
| Pass Score | 700 / 1000 |
| Duration | 100 minutes |
| Questions | ~40-60 |
| Cost | $165 USD (varies by region) |
| Provider | Pearson VUE |
| Validity | Renew annually (free via Microsoft Learn) |
| Prerequisites | None enforced; AZ-900 + AI-900 background recommended |
| Question Types | Multiple choice, Multiple response, Drag-and-drop, Case study |
| Replaces | AZ-204 (retiring 31 July 2026) |
| Official Page | Microsoft Learn — AI-200 |
Skills Measured
The official Microsoft AI-200 study guide lists 4 domains. Focus your study time using the weights below.
1. Develop containerised solutions on Azure (20–25%)
Implement container application hosting
- Build, store, version, and manage container images using Azure Container Registry
- Build and run images using ACR Tasks
- Deploy containers to Azure App Service, including configuring App Service to supply environment variables and secrets
Implement container-orchestrated solutions
- Deploy applications to Azure Container Apps, including environment configuration and revision management
- Implement event-driven scaling using KEDA in Container Apps
- Deploy and manage applications to Azure Kubernetes Service (AKS) using manifest files
- Monitor and troubleshoot solutions on AKS and Container Apps by inspecting logs, events, and end-to-end connectivity
2. Develop AI solutions using Azure data management services (25–30%)
Develop AI solutions with Azure Cosmos DB for NoSQL
- Connect to Azure Cosmos DB for NoSQL using the SDK and run queries
- Optimise query performance and Request Units (RUs) using indexing policies and consistency levels
- Store and retrieve embeddings and execute vector similarity search for semantic retrieval
- Implement a change feed processor to detect and handle new or updated items
Develop AI solutions with Azure Database for PostgreSQL
- Connect and query Azure Database for PostgreSQL using SDKs
- Model schemas and implement indexing strategies
- Implement indexing strategies including optimising query latency and reducing pgvector compute overhead
- Configure compute, memory, and storage resources to support vector workloads
- Run vector similarity search, including storing embeddings, semantic retrieval, and implementing RAG patterns with metadata filters
- Implement connection optimisation to improve throughput and minimise latency
Integrate Azure Managed Redis in AI solutions
- Implement Azure Managed Redis data operations including caching, expiration, and invalidation
- Implement vector indexing to enable similarity search
3. Connect to and consume Azure services (20–25%)
Develop event- and message-based AI solutions
- Queue and process back-end operations using Azure Service Bus, including dead-letter queue handling, messages, topics, and subscriptions
- Implement event-driven workflows using Azure Event Grid, including filters, custom events, and retries
Develop and implement Azure Functions
- Build serverless APIs, including implementing triggers and bindings
- Configure and deploy function apps
4. Secure, monitor, and troubleshoot Azure solutions (20–25%)
Implement secure Azure solutions
- Secure secrets using Azure Key Vault, including rotation and retrieval
- Store and retrieve app configuration information using Azure App Configuration
Monitor and troubleshoot Azure solutions
- Trace distributed systems using OpenTelemetry SDKs
- Write KQL queries to analyse logs and metrics
Our Free 27-Module Study Guide
We’ve built a complete free interactive study guide on the Guided platform that maps every exam objective to a focused module. Each module includes:
- ELI5 toggle — analogy first, then technical depth
- CompareTable components for “X vs Y” decisions (which is half the exam)
- Exam tips in expandable disclosures
- Real-world scenarios featuring four recurring back-end developers
- Flashcards for key terms
- 3 quiz questions per module with full “why wrong” explanations
Module Breakdown
| Domain | Modules |
|---|---|
| 1 — Containers | 8 modules: ACR · ACR Tasks · App Service · Container Apps · KEDA · AKS · troubleshooting |
| 2 — Data services | 8 modules: Cosmos NoSQL (incl. vectors + change feed) · PostgreSQL + pgvector · RAG · tuning · Managed Redis |
| 3 — Connect Azure | 6 modules: Service Bus queues · topics · Event Grid · Functions · deployment · decision module |
| 4 — Secure / monitor | 5 modules: Key Vault · App Configuration · OpenTelemetry · KQL · end-to-end observability |
| Total | 27 free modules |
Start the AI-200 study guide →
Practice Exam — Coming Soon
A 250-question practice exam (4 domain banks, exam-mode + study-mode) is in production. Bookmark this page or follow @aguidetocloud for the launch.
Microsoft Certification Path
Microsoft Azure follows three levels: Fundamentals → Associate → Expert. AI-200 sits at the Associate level alongside AI-103 (AI engineering) and AZ-204 (retiring).
Related Certifications
If you’re studying for AI-200, these are good companions:
- AI-103: Azure AI App and Agent Developer — pair with AI-200 for the full AI-engineering picture
- AI-901: Azure AI Fundamentals — start here if you’re new to AI on Azure
- AZ-204: Developing Solutions for Microsoft Azure — the exam AI-200 replaces (retiring 31 July 2026)
- AI-300: MLOps Engineer Associate — if your career path leads toward ML platforms
Study Tips
- Lean on the free 27 modules. Every exam objective has a home module. The “decision” modules (Choosing the Right Messaging Service, end-to-end observability) are where exam questions love to hide.
- Don’t skip Domain 2 — it’s the heaviest at 25-30%. Vector search across Cosmos, pgvector, and Managed Redis is the most distinctive content on this exam.
- Get hands-on with managed identity. Almost every “what’s the right pattern?” answer that mentions a connection string is a wrong answer. Practice the Microsoft Entra + managed identity flow against Service Bus, ACR, Cosmos, and Key Vault.
- Read the question for the data path. AI-200 scenarios often hinge on “where does the data live?” Map the data path before picking a service — the right Azure primitive almost always falls out.
- Use the official study guide as the spine. Microsoft’s study guide is the source of truth for what’s tested.
Quick Links
Frequently asked questions
What people ask me most about AI-200 — the exam quietly replacing AZ-204 on 31 July.
Is AI-200 worth taking in 2026? #
What's the difference between AI-200 and AI-103? #
How long do I need to prepare for AI-200? #
Do I need AZ-204 before AI-200? #
Is there a practice exam for AI-200? #
Compare AI-200 across AWS & Google Cloud → Cert Compass
Frequently Asked Questions
1. Is AI-200 worth taking in 2026?
If you were planning to take AZ-204, yes — AI-200 is the exam that replaces it on 31 July 2026. Same Associate level, same $165, but the content is rebuilt around what Azure developers actually do today: containers, vector data, messaging, observability. If you sit AZ-204 now and need to renew later, you'll renew into AI-200 anyway. Save the round trip — start on AI-200.
2. What's the difference between AI-200 and AI-103?
AI-103 tests you on the AI itself — model selection, agents, RAG design, prompt flows. AI-200 tests you on the back end that runs AI in production — containers, Cosmos DB with vectors, pgvector, Service Bus, Key Vault, KQL. AI-103 is for the person designing the assistant. AI-200 is for the person wiring it into a real Azure environment. Both can sit next to each other on a CV — they don't overlap much.
3. How long do I need to prepare for AI-200?
Four to eight weeks of part-time study if you already write Azure back-end code. Longer if vector databases or KEDA / Container Apps are new to you. Domain 2 (data services with embeddings) is the heaviest at 25–30% — budget extra time for Cosmos vector search, pgvector RAG, and Managed Redis. Our [free 27-module guide](/guided/ai-200/) maps every objective to a focused module so you can pace yourself.
4. Do I need AZ-204 before AI-200?
No — and you probably shouldn't take AZ-204 at this point. Microsoft is retiring AZ-204 on 31 July 2026, so anything you certify on now has a short shelf life. The recommended background is Azure SDK familiarity, comfort with containers, and basic Python. AZ-900 + AI-900 give you the concept layer for free, but neither is enforced. Start on the [free study guide](/guided/ai-200/) and see where the gaps are.
5. Is there a practice exam for AI-200?
Not yet — a 250-question practice exam covering all four domains is in production. Until it ships, lean on Microsoft's free [practice assessment](https://learn.microsoft.com/en-us/credentials/certifications/exams/ai-200/) and the quiz questions inside each of the [27 free modules](/guided/ai-200/) — three questions per module with full explanations on every wrong answer, which is what most paid practice exams charge for.