AI-901: Microsoft Azure AI Course with Hands On Simulations

Learn to access and complete hands-on simulations via assignments hosted on ai.azure.com (Microsoft Foundry), including opening links in a new tab, submission, and troubleshooting a Udemy checkmark issue.

Create a free Azure account to access $200 credit for 30 days and many free services for 12 months, then manage billing and login via portal.azure.com.

Install python on Windows by downloading the standalone installer from python.org, enabling admin privileges, adding python.exe to the path, and verifying python and pip versions.

Install Visual Studio on Windows with system installer x64, accept license, then launch the IDE, create a test code folder, add TestCode.py, and install the Python extension for IntelliSense.

Explore Microsoft Foundry as an AI studio for creating projects and adding models while navigating frequent UI changes, classic vs foundry views, the model catalog, and Azure resource groups.

Discover how to redo simulations after finishing an assignment on Udemy. Follow the steps: go to summary, back to assignment, then instructions, to access the simulation link anytime.

Explore responsible AI fundamentals with a focus on fairness and inclusiveness, and learn how Azure's fairness assessment in the AI dashboard helps ensure equal treatment across demographics.

Evaluate reliability and safety in an AI solution using the Responsible AI dashboard. Analyze error distributions to identify high-risk cohorts and reinforce trust.

Explore privacy and security in Azure AI, covering compliance policies, access controls, encryption in transit and at rest, auditing, and tools like SmartNoise and Counterfeit for regulatory and resilience testing.

Explain transparency in AI by communicating how outputs affect decisions for people, stakeholders, and operators, using global, local, and model explanations in Azure Machine Learning's responsible AI dashboard.

Enable accountability in AI through MLOps practices with human oversight, logging, monitoring, and alerts across the end-to-end ML lifecycle from development to deployment.

Describe how generative AI creates new content using deep learning and vast training data, powered by transformer architectures, prompts, and token-based outputs, while noting potential hallucinations and math limitations.

Identify the appropriate model deployment options and configuration parameters in Microsoft Foundry via ai.azure.com, choosing chat completion or image generation and considering global standard vs data zone standard.

Explore common AI workloads, from generative and agentic AI to text analysis, speech, computer vision, and information extraction, with practical scenarios like chatbots, summaries, personalization, and automation.

Explore the features of speech recognition and speech synthesis, including real-time and batch transcription, language detection, custom vocabulary, noise handling, and natural neural voice text-to-speech.

Explore computer vision and image generation capabilities, including image classification, object detection, facial detection, OCR, image tagging, and text-to-image creation with configurable prompts and variations.

Identify techniques to extract information from text, images, audio, and videos, including keyword extraction, entity recognition, sentiment analysis, speech recognition, and object detection for actionable insights.

Learn to craft effective system and user prompts for generative AI by mastering clarity, context, and structure, and using examples and constraints to guide precise, consistent responses.

Deploy and manage models in microsoft foundry by selecting a model from ai.azure.com, reviewing capabilities, use cases, and pricing, then create and deploy a new project.

deploy a simple agent with Microsoft Foundry to learn an employee handbook from a pdf and answer time-off and smoking policy questions via a GPT model.

Learn to create a lightweight client application for an agent using curl to send JSON requests to an AI model endpoint, including the API key, with minimal code.

Develop a lightweight Python app for text analysis through document translation using Azure Translator, setting source and target languages, uploading the file, and saving the translated output.

Learn how a deployed multimodal model interprets visual input within prompts by combining images with text to describe scenes, identify objects, and generate accessible summaries.

Explore creating visual outputs with generative models in Microsoft Foundry on ai.azure.com. Deploy Flux 0.2 Pro for text-to-image and image-to-image prompts, then refine visuals with a cat riding a motorcycle.

Deploy an image generation model on Azure with Foundry and vision capabilities. Write Python code to send prompts, decode base64 results, and save the generated image as a PNG.

Explore how Azure Content Understanding extracts structured data from documents and forms using OCR, layout, and document fields, enabling apps and workflows with confidence scores.

Learn to use Azure Content Understanding in Foundry to extract information from a PDF with OCR, and capture service name, platform, and key benefits in text and JSON results.

Extract and analyze audio and video content with Azure Content Understanding to identify spoken words, summarize conversations, detect topics, and output structured data for easy search, review, and workflow integration.

Build a lightweight application that uses Azure content understanding to extract audio information. Wire Python script with endpoint and key, choose a call center analyzer, and output JSON results.