Azure AI Cloud Developer Associate (AI-200) Learning Lab

Welcome to the hands-on learning workspace for Microsoft Certified: Azure AI Cloud Developer Associate (AI-200).

This repository is organized into progressive, code-first modules designed to build real-world, cloud-native AI systems on Microsoft Azure.

---

🗺️ Learning Roadmap & Exam Domains

flowchart TD
    M1["01: Identity, Auth & Model Inference\n(DefaultAzureCredential, AI Foundry SDK)"] --> M2["02: Vector Search & Data Ingestion\n(Azure AI Search, Cosmos DB Vector Index)"]
    M2 --> M3["03: Cloud-Native AI Compute\n(FastAPI, Docker, Azure Container Apps)"]
    M3 --> M4["04: Async & Event-Driven AI\n(Service Bus, Event Grid)"]
    M4 --> M5["05: Observability & Evaluation\n(App Insights, OpenTelemetry, AI Guardrails)"]

Loading

---

📂 Module Breakdown

Module	Core Topics	Key Azure Services / SDKs
01-ai-foundry-identity-inference	Zero-trust auth, Entra ID, chat completions, streaming, retry policies	azure-identity, azure-ai-inference, openai
02-vector-search-and-data	Embeddings, vector indexing, hybrid search (BM25 + vector), RAG	azure-search-documents, azure-cosmos
03-container-apps-and-compute	Microservice AI APIs, ACA scaling, health probes, Docker	Azure Container Apps, Azure Container Registry, FastAPI
04-event-driven-ai-pipelines	Async batch processing, message queuing, dead-lettering	Azure Service Bus, Event Grid, Storage Queues
05-observability-and-evaluation	Distributed tracing, token tracking, latency, groundness evaluation	Azure Monitor, Application Insights, OpenTelemetry

---

🔑 Key Cloud AI Principle: Zero-Trust & Passwordless

In AI-200, hardcoded API keys are considered an anti-pattern. Production backends use Microsoft Entra ID Role-Based Access Control (RBAC) and Managed Identities (DefaultAzureCredential) so secrets never leak into code or environment files.