7 day study plan and cheatsheet for Microsoft Azure AI Fundamentals (AI-900)

Exam Syllabus & Weightings

According to Microsoft’s May 2, 2025 update, the key areas and their weightings are:

• Describe AI workloads and responsible AI (15–20%)
• Fundamental principles of machine learning on Azure (20–25%)
• Computer vision workloads on Azure (15–20%)
• Natural Language Processing (NLP) workloads on Azure (15–20%)
• Generative AI workloads on Azure (15–20%) Microsoft Learnwhizlabs.com

Microsoft Azure Artificial Intelligence Fundamentals (AI-900)

Key topics under each domain (from detailed Microsoft syllabus):

• AI Workloads & Responsible AI: anomaly detection, vision, NLP, knowledge mining; principles like fairness, reliability, privacy, inclusiveness, transparency, accountability
• Machine Learning Fundamentals: regression, classification, clustering; dataset features and labels; training vs validation; Visual tools like AutoML and ML designer in Azure ML
• Computer Vision: image classification, object detection, OCR, face recognition; services like Computer Vision, Custom Vision, Face, Form Recognizer
• NLP: key phrase extraction, entity recognition, sentiment analysis, language modeling, speech (recognition/synthesis), translation; Azure Language, Speech, Translator services; conversational AI via bots, Power Virtual Agents, Azure Bot Service
• Generative AI: Recognize generative AI solutions and Azure OpenAI Service capabilities

Cheat sheet of concepts and Azure services that appear in the AI-900 (Azure AI Fundamentals) exam

1. AI Concepts

• Artificial Intelligence (AI) — Systems that simulate human intelligence to perform tasks; core of what the exam is about.
• Machine Learning (ML) — Algorithms that learn patterns from data; key part of Azure AI workloads.
• Deep Learning — ML using neural networks with many layers; used in vision, NLP, generative AI.
• Supervised Learning — ML with labeled data; common in Azure ML demos.
• Unsupervised Learning — ML without labels; includes clustering scenarios.
• Classification — Predict categories (e.g., spam vs not spam); Azure ML use case.
• Regression — Predict continuous values; tested in ML fundamentals questions.
• Clustering — Group data without predefined labels; shows up in ML concept questions.
• Responsible AI Principles — Fairness, reliability, privacy, inclusiveness, transparency, accountability; directly tested.

2. Azure ML & Data Services

• Azure Machine Learning — Cloud platform to build/train/deploy ML models; central ML tool for AI-900.
• Azure ML Designer — Drag-and-drop interface for ML workflows; used in AI-900 labs.
• Automated ML (AutoML) — Automatically trains and tunes models; exam will test concept, not coding.
• Data Labeling in Azure ML — Tagging data for supervised learning; important for dataset preparation.

3. Computer Vision Services

• Azure Computer Vision — Extracts info from images (OCR, tags, descriptions); key for vision questions.
• Custom Vision — Train custom image classifiers; AI-900 covers workflow basics.
• Face API — Detects and identifies faces; appears in scenario questions.
• Form Recognizer — Extracts structured data from documents/forms; common exam example.

4. NLP & Speech Services

• Azure Language Service — Unified NLP capabilities like sentiment analysis, key phrases, entity recognition; AI-900 staple.
• Text Analytics — Older name for parts of Language Service; know it for historical references.
• Translator — Real-time text translation; common scenario in AI-900.
• Azure Speech Service — Speech-to-text, text-to-speech, speech translation; often tested.
• Conversational Language Understanding (CLU) — Builds intent/utterance models for chatbots.
• Azure Bot Service — Framework for building chatbots; tested in NLP + conversational AI topics.
• Power Virtual Agents — No-code bot creation; exam may reference for low-code AI scenarios.

5. Generative AI Services

• Azure OpenAI Service — Access to GPT models (text generation, summarization, Q&A); newest AI-900 topic.
• Prompt Engineering — Designing inputs to guide AI output; conceptual understanding tested.

6. General Azure Integration Services

• Azure Cognitive Services — Suite of pre-built AI APIs (Vision, Language, Speech, Decision); umbrella category for many above.
• Knowledge Mining with Azure Cognitive Search — Uses AI to extract and search structured/unstructured data; may appear in “AI workloads” section.

Study Plan to crack AI-900 in 7 Days

Days 1–2: Build Your Foundation

• Use Microsoft Learn modules tied to AI-900 for self-paced learning. Focus on AI workloads, responsible AI principles, and ML basics.
• Skim through study guides — start with Microsoft’s study guide and the codecademy certification path outline Microsoft LearnCodecademy.

Days 3–4: Deep Dive Into Workloads

• Computer Vision: Understand scenarios + Azure services (Custom Vision, OCR).
• NLP & Bots: Learn Azure services and common use cases.

Day 5: Generative AI + Practice

• Explore Azure OpenAI Service overview.
• Begin practice tests — Microsoft’s free sample assessment or other platforms like Whizlabs or Udemy-style mocks whizlabs.com.

Days 6–7: Review & Reinforce

• Focus on weaker areas revealed by practice tests.
• Use flashcards or mind maps for quick recall (highly recommended by peers)
• “Flashcards on Quizlet: … extremely helpful for reviewing key concepts…”
• Join Reddit or forum threads for last-minute peer tips and encouragement

Exam Day Tactics

• Use the Exam Sandbox: Get comfortable with layout and question formats ahead of time Microsoft Learn.
• Testing Strategy: Flag unfamiliar questions and return later. Manage your pace.
• Stay Calm: Confidence comes from practice. A peer with zero AI background scored 850 after focused study Reddit.

Good luck with your prep!!