AI-901: Microsoft Azure AI Fundamentals for Business Professionals

Teaching lesson: Microsoft's six Responsible AI principles (fairness, reliability & safety, privacy & security, inclusiveness, transparency, accountability) and how they show up in Azure AI services and governance features.

Hands-on lab (portal only): configure Content Safety, examine default guardrails in Foundry model deployments, and walk through Azure AI resource governance controls.

Teaching lesson: what large language models, small language models, and multimodal models are; tokens, context windows, temperature, top-p, and how model choice and configuration affect output quality, cost, and latency.

Hands-on lab (portal only): browse the Microsoft Foundry model catalog, compare GPT, Phi, and open-source models, examine model cards, and review deployment options in the Foundry portal.

Teaching lesson: the four Azure AI workload categories (generative AI, natural language, computer vision, document intelligence) and which Microsoft Foundry tools and Azure AI services map to each.

Hands-on lab (portal only): tour every Azure AI workload area inside Azure AI Foundry — Language Playground (sentiment, key phrases, NER, PII, summarization), Speech AI Services (transcription, voice synthesis, speech translation), Vision AI Services (image captions, tags, object detection, OCR), Images Playground (gpt-image-1 image generation), and Content Understanding (structured document extraction). No code required — everything runs in the browser against pre-provisioned resources.

Teaching lesson: prompt structure (system, user, assistant), few-shot prompting, grounding, chain-of-thought, and how prompt design drives output quality. Uses narrative examples rather than code.

Hands-on lab (portal only): deploy a chat model in Microsoft Foundry and iterate on prompts in the Foundry playground — system prompts, few-shot examples, temperature, and grounding with your own data.

Teaching lesson: model deployment options (serverless vs dedicated, PTU vs standard), deployment regions and quotas, and how to pick the right deployment for a business scenario.

Hands-on lab (portal only): create a Foundry project, deploy GPT-4o-mini, and hold a multi-turn conversation in the Foundry chat playground — adjust system prompt, temperature, and max tokens.

Teaching lesson: what an agent is (model + instructions + tools + knowledge), single-agent patterns, Foundry Agent Service, tools (functions, file search, code interpreter), and when to build an agent vs. a plain chat experience.

Hands-on lab (portal only): create a single-agent solution in Microsoft Foundry — configure the model, system instructions, attach file-search knowledge, and chat with the agent in the playground.

Teaching lesson: natural-language capabilities in Azure AI — sentiment, key phrase extraction, entity recognition, PII detection, summarization, translation — and how Foundry and Azure AI Language differ.

Hands-on lab (portal only): use a pre-provisioned Azure AI Services resource to run five language analysis capabilities in Azure AI Foundry — sentiment analysis with opinion mining, key phrase extraction, named entity recognition, PII detection, and extractive summarization — against sample Nimbus Trail customer reviews. No code required.

Teaching lesson: speech-to-text, text-to-speech, neural voices, real-time vs batch transcription, speech translation, and the Foundry/Speech Studio tools that deliver these capabilities.

Hands-on lab (portal only): explore Azure AI Speech capabilities using a pre-provisioned Azure AI Services resource in Azure AI Foundry — transcribe live speech with real-time speech-to-text, audition and compare neural voices with text-to-speech, and translate spoken English into Spanish or French with Speech Translation. No code required.

Teaching lesson: image analysis (tags, captions, OCR, object detection, face detection), image generation with DALL-E / GPT-Image, and how Vision Studio and Foundry handle each capability.

Hands-on lab (portal only): use Azure AI Foundry to analyze images with captions, tags, object detection, and OCR via the Vision AI Services section, then generate images with gpt-image-1 in the Images Playground — all from a single portal.

Teaching lesson: Azure AI Content Understanding — extracting structured data from documents, images, audio, and video using analyzers and schemas; comparison with Document Intelligence.

Hands-on lab (portal only): use Azure AI Foundry to create a Content Understanding analyzer, define an extraction schema for invoice-style documents, and run structured data extraction against a sample invoice — no code required.

Teaching lesson: how a realistic Azure AI solution combines Foundry models, agents, tools, speech, vision, and Content Understanding — exam-style scenario walkthroughs and architecture patterns.