Episode Details

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

AI-900 has been replaced by AI-901, and the new Azure AI Fundamentals exam is different: 55–60% of it is now hands-on work in Microsoft Foundry (models, agents, speech, vision and Content Understanding). In this episode: • What AI-901 measures and how it differs from AI-900 • The six responsible AI principles, with examples • A plain-language tour of Microsoft Foundry for the exam • 10 practice questions with explanations • A 2-week study plan Chapters: 00:00 AI-900 is replaced by AI-901: what changed 02:30 Who the exam is for 04:35 The two domains, weights and pass mark 06:15 Responsible AI: the six principles and their keywords 11:35 Tokens, context window, temperature and top-p 17:30 Choosing and deploying the right model 20:40 AI workloads: generative, agentic, text, speech, vision, extraction 22:55 Domain 2: building in Microsoft Foundry 24:35 Keyless authentication, system vs user prompts 27:00 A single agent: define, connect, test, deploy 28:15 Hallucinations, grounding and RAG 31:15 Azure Content Understanding: extract, classify, generate 35:00 10 practice questions with explanations 46:15 Two-week study plan 49:05 Extra time if the exam isn't in your language 51:40 More practice: the AZ AI app Practise more AI-901 questions, with every answer explained: AZ AI – Cert Exam Prep (iPhone, iPad): https://apps.apple.com/app/id1611614573 AI & Machine Learning Tutor (41 AI certifications): https://apps.apple.com/app/id1611593573 Official sources: Microsoft Learn certification page: https://learn.microsoft.com/en-us/credentials/certifications/azure-ai-fundamentals/ AI-901 study guide: https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/ai-901 Not affiliated with or endorsed by Microsoft. Azure and Microsoft Foundry are trademarks of Microsoft.