Course overview
Move AI prototypes toward production-ready Azure solutions. This Express course helps developers design containerized, serverless, event-driven, vector-enabled, observable cloud backends for AI apps and agents using Azure services such as Container Apps, App Service, AKS, Functions, Service Bus, Event Grid, Cosmos DB, PostgreSQL with pgvector, Redis, Key Vault, App Configuration, and Azure Monitor.
A production cloud developer course that combines AI-200 cloud solution topics with selected AI-103 and AI-3026 integration patterns. It focuses on the Azure application layer that hosts, connects, secures, scales, and observes AI apps and agents.
Who this is for
Cloud developers, backend developers, AI application developers, solution architects, and engineers responsible for productionizing AI-driven applications on Azure.
Prerequisites
Azure developer experience, basic container and API concepts, and familiarity with at least one programming language. Prior AI app experience recommended.
Recommended prior course or experience: Azure AI Developer Express for AI app and agent logic; Azure fundamentals or AZ-204-style development experience for Azure newcomers.
Course outline
Azure AI cloud developer landscape
Backend and cloud roles for AI applications
How compute, data, messaging, and identity fit
AI logic vs cloud solution development
Reference architecture for AI apps and agents
Containerized compute for AI solutions
Build, store, version, and manage images
Azure Container Registry and build automation
Deploy to App Service and Container Apps
Environment variables, secrets, and config
Container orchestration and scaling
Container Apps environments, revisions, KEDA
AKS orientation for teams needing Kubernetes
Monitoring and troubleshooting containers
Reliability: probes, retries, and rate limits
Azure Functions and serverless APIs
HTTP-triggered functions and lightweight APIs
Triggers and bindings for queues and events
Functions vs containers decision matrix
Observability for serverless workflows
Messaging, eventing, and service integration
Service Bus queues, topics, and dead-letter
Event Grid for loosely coupled workflows
Retries, poison messages, and fan-out
Async AI workflows for docs and agent tasks
Operational and vector data services
Cosmos DB for NoSQL in AI applications
Embeddings, vector search, and change feed
Azure Database for PostgreSQL with pgvector
Chunking, indexing, cost, and retrieval perf
Caching and application performance
Azure Managed Redis for cache and speedup
Cache invalidation and AI response latency
Cost and performance tradeoffs
Protecting cached data and secrets
Configuration, secrets, and secure access
Key Vault, managed identities, secret rotation
App Configuration for flags and settings
Environment separation and deploy readiness
Handoff to Secure AI and Cloud Workloads Express
Monitoring, observability, and troubleshooting
Azure Monitor, App Insights, logs, and KQL
OpenTelemetry and distributed tracing
Correlation IDs across AI calls and queues
Incident runbooks and operational diagnostics
Production-readiness workshop
Map a solution to hosting, data, and security
Create a deployment and troubleshooting list
Identify handoffs for security, SOC, and data
