Explanation:

The correct answer is "adding and connecting modules on a visual canvas." According to the Microsoft Azure AI Fundamentals (AI-900) Official Study Guide and the Microsoft Learn module "Explore automated machine learning in Azure Machine Learning," the Azure Machine Learning designer is a drag-and-drop, no-code environment that allows users to create, train, and deploy machine learning models visually. It is specifically designed for users who prefer an intuitive graphical interface rather than writing extensive code.
Microsoft Learn defines Azure Machine Learning designer as a tool that allows you to "build, test, and deploy machine learning models by dragging and connecting pre-built modules on a visual canvas." These modules can represent data inputs, transformations, training algorithms, and evaluation processes. By linking them together, users can create an end-to-end machine learning pipeline.
The designer simplifies the machine learning workflow by allowing data scientists, analysts, and even non- developers to:
* Import and prepare datasets visually.
* Choose and connect algorithm modules (e.g., classification, regression, clustering).
* Train and evaluate models interactively.
* Publish inference pipelines as web services for prediction.
Let's analyze the other options:
* Automatically performing common data preparation tasks - This describes Automated ML (AutoML), not the Designer.
* Automatically selecting an algorithm to build the most accurate model - Also a characteristic of AutoML, where the system tests multiple algorithms automatically.
* Using a code-first notebook experience - This describes the Azure Machine Learning notebooks environment, which uses Python and SDKs, not the Designer interface.
Therefore, based on the official AI-900 learning objectives and Microsoft Learn documentation, the Azure Machine Learning designer allows you to create models by adding and connecting modules on a visual canvas, providing a no-code, interactive experience ideal for users building custom machine learning workflows visually.