21 docs tagged with "ai-900"

Understand what Azure Machine Learning is, its key components, and why it is the platform of choice for building custom ML solutions on Azure.

Understand how machines interpret visual information — the core Computer Vision tasks, how CNNs work conceptually, and the critical distinctions that AI-900 tests most often.

Microsoft's 6 Responsible AI principles — definition, real-world examples, failure cases, and the unavoidable trade-offs between principles when they conflict.

Understand what NLP is, the core tasks it enables, and how an NLP pipeline transforms raw text into structured insights — the conceptual foundation for AI-900 language workloads.

Understand what Artificial Intelligence is, why it matters, and how it differs from traditional rule-based programming — the starting point for the AI-900 exam.

Understand what Azure AI is, why organizations use managed AI services instead of building from scratch, and how Azure AI Foundry serves as the unified development platform.

Understand what Generative AI is, how Large Language Models work, and the fundamental shift from analytical AI to content-creating AI — the most rapidly evolving area in AI-900.

Understand what Machine Learning is, why it matters, and the key distinction between supervised and unsupervised learning — the conceptual foundation for AI-900 exam domain 2.

Understand the containment hierarchy AI ⊃ ML ⊃ DL with definitions, trade-offs, and a practical decision framework.

A detailed reference for all seven Azure AI prebuilt services covered in AI-900 — capabilities, use cases, and key distinctions between services.

Explore the three development approaches in Azure ML — AutoML, Designer, and Notebooks — and understand the responsible AI tools built into the platform.

Deep dive into Azure AI Language and Azure AI Speech — capabilities, Conversational Language Understanding, custom models, and the exam-critical service selection scenarios for NLP workloads.

Master Azure OpenAI Service — available models, prompt engineering techniques, the RAG pattern, and Azure AI Foundry integration for building production generative AI applications.

Azure tools that operationalize Responsible AI — the Responsible AI Dashboard in Azure ML, Azure AI Content Safety, Transparency Notes, and how to embed RAI practices across the full AI lifecycle.

Azure Vision services in depth — Azure AI Vision, Custom Vision, Face API, and Document Intelligence — with capability tables, industry use cases, and exam-critical service selection scenarios.

Master the essential ML vocabulary tested in AI-900 — features, labels, overfitting, underfitting, model evaluation metrics, and the train/validation/test split.

Real-world case studies applying RAI principles across industries, plus a comprehensive AI-900 exam preparation guide covering all 8 workshops with key concepts and service selection patterns.

A systematic decision framework for selecting the right Azure AI service given a business problem — the most exam-critical skill in AI-900 Workshop 2.

Explore the four primary AI workload categories in the AI-900 exam: NLP, Computer Vision, Speech, and Generative AI — with definitions, capabilities, Azure service mappings, and a business problem decision table.

Understand the unique Responsible AI risks of generative systems — hallucination, bias amplification, copyright, and deepfakes — and the Azure safety stack that mitigates them.

Understand the end-to-end Machine Learning lifecycle — from data collection to deployment — and the iterative nature of real-world ML projects.