
Explore how artificial intelligence sits on the Azure cloud, a beginner gateway to AI in Azure, covering generative AI, AI workloads, and building agents with Microsoft Foundry.
Define artificial intelligence, its capabilities, and the ethical principles of responsible AI. Explore six AI workloads and the four pillars—predict, decide, converse, see—plus azure Foundry integration.
Learn six AI workloads—text and language, computer speech, computer vision, information extraction, generative AI, and agents—and use machine learning with Azure language service and content understanding for real-time use cases.
Learn the six principles of responsible AI: fairness, reliability and safety, privacy and security, inclusiveness, transparency, and accountability, and how governance and thorough testing protect AI applications.
Explore AI agents through a browser-based model, learning how local hardware affects inferencing, model choices, and how chatbots like ChatGPT operate as stateless, generative AI.
Learn how Azure combines AI with a layered model—tenant, subscription, resource group, and resources—using Entra ID for identity, with pay as you go billing and services from IaaS to SaaS.
Explore generative AI and agents, covering large and small language models, tokens, embeddings, attention, and multi-modal capabilities, with Azure implementations and reinforcement learning.
Explore how gen AI models convert text into tokens and embeddings, then apply attention in a transformer to predict the next word. See how tokens, embeddings, and attention drive generation.
Explore what constitutes an artificial intelligence agent, including its brain (model), system prompts, tasks, and tools, and how grounding data and retrieval augmented generation enable autonomous, multi-agent collaboration.
Structure prompts with a system message and persona, ground outputs with explicit context and data, and define a specific task to improve prompt engineering outcomes.
Explore how AI agents differ from standalone models, and why grounding sources steer responses in structured output. Learn choosing small language models for devices with limited connectivity in Azure.
Explores deploying models in Microsoft Foundry (ai.azure.com) with GPT 5.4, using the playground, system prompts, and model router, and highlights Entra ID authentication for Azure AI tasks.
Create autonomous agents in Foundry by configuring models, instructions, tools, and knowledge bases; attach tools such as web search and knowledge sources, and deploy agents via the Foundry portal.
Explore natural language processing concepts, from key phrase extraction and named entity recognition to sentiment analysis and summarization, and compare general AI models with Azure language services.
Explore text preprocessing for natural language processing, including tokenization, normalization, stop word removal, stemming and lemmatization, and part-of-speech tagging to prepare raw text for AI analysis.
Contrast statistical text analysis with semantic models, tracing NLP from word counting to embedding-based meaning spaces that fuel large language models and transformer-era approaches.
Compare general purpose AI models with Azure language for text analytics, highlighting deterministic structured outputs, confidence scores, and capabilities like language detection, named entity recognition, and PII redaction.
Learn how to use Azure language in Foundry to provision a language service or a Foundry resource, access endpoints, and perform sentiment, named entity recognition, and personally identifiable information detection.
Explore the speech recognition and speech synthesis pipelines, including audio to text and text to audio, with MFCC features, acoustic and language modeling, normalization, and prosody.
Explore Azure Speech in foundry tools, including speech to text and text to speech, via the playground and Speech SDKs, plus Voice Live for conversations with a speech capable agent.
Learn core computer vision concepts, including image classification, object detection with bounding boxes, semantic segmentation, and contextual image analysis, plus image-to-text and text-to-image workflows, with Microsoft Azure Foundry hands-on.
Compare CNNs and vision transformers, noting CNNs' local feature extraction and vision transformers' global context via patch embeddings and attention, plus diffusion and multimodal models.
Explore computer vision in Azure, using multi-modal models to describe images, extract text, compare images, and deploy GPT-4 models in Foundry with image URL or base64 inputs.
Explore image and video generation in Azure AI by leveraging text-to-image inferencing, prompts, and foundry models like SORA, along with asynchronous, multi-modal workflows.
Convert unstructured data from documents, images, audio, and video into structured data using information extraction, OCR, and field mapping with the Azure content understanding service.
Learn how Azure content understanding enables information extraction from documents, audio, and video with predefined and custom analyzers, REST API access, and automated invoice processing workflows.
Explore content understanding in the foundry portal with predefined analyzers like OCR and form extraction. Learn to deploy models and handle asynchronous JSON outputs with confidence scores.
Prepare for the Microsoft Azure AI Fundamentals (AI-901) certification with a focused course covering the AI concepts, Azure AI workloads, and Microsoft Foundry capabilities you need to understand for the current exam.
AI-901 is designed for learners beginning their journey in AI solution development. The current exam combines foundational AI concepts with practical implementation using Microsoft Foundry, including generative AI applications and agents, text and speech solutions, computer vision, image generation, and information extraction. This course is structured around those core areas.
You will begin with the fundamentals of Artificial Intelligence and Azure AI, including common AI workloads and the principles of Responsible AI. You will also become familiar with Microsoft Azure and Microsoft Foundry before moving into modern Generative AI concepts and AI agents.
The Generative AI and Agents section introduces how generative AI works, key concepts behind modern AI models, AI agents, prompt considerations, model deployment in Microsoft Foundry, and creating agents in Foundry.
You will then explore Natural Language Processing and Azure AI Language, including text preprocessing, statistical and semantic approaches, text analytics, and Azure Language capabilities in Microsoft Foundry.
The course also covers Azure AI Speech, including speech-to-text and text-to-speech concepts, Azure Speech services, and speech capabilities available through Foundry.
In the Computer Vision section, you will learn fundamental computer vision concepts, explore Convolutional Neural Networks and Vision Transformers, understand computer vision capabilities in Azure, and explore image and video generation.
The final section focuses on information extraction. You will learn the fundamentals of extracting information from documents and explore Azure Content Understanding in Microsoft Foundry.
Throughout the course, the focus is on understanding what each Azure AI capability does, where it fits, and how it is used within modern AI solutions.
By completing this course, you will have a structured understanding of the major AI concepts and Azure AI workloads covered by AI-901, along with the Microsoft Foundry capabilities used to implement modern AI solutions.
Whether you are preparing specifically for AI-901, beginning your Azure AI journey, or looking to understand modern AI technologies such as Generative AI and AI agents, this course provides a structured foundation for your next step in Azure AI.