Components of an AI application

  • 3 minutes

Microsoft supports each layer of an AI application: the data layer, model layer, compute layer, and orchestration layer.

Data Layer

The data layer is the foundation of any AI application. It includes the collection, storage, and management of data used for training, inference, and decision-making. Common data sources include structured databases such as Azure SQL and PostgreSQL, unstructured data, such as documents and images, and real-time streams. Azure services like Cosmos DB and Azure Data Lake are often used to store and manage large-scale datasets efficiently.

Microsoft offers databases as a Platform-as-a-Service (PaaS). Platform services are managed cloud services that provide the foundational building blocks for developing, deploying, and running applications without requiring users to manage the underlying infrastructure. PaaS sits between Infrastructure-as-a-Service (IaaS) and Software-as-a-Service (SaaS) in the cloud service model.

Model Layer

The model layer involves the selection, training, and deployment of machine learning or AI models. Models can be pretrained (for example: Azure OpenAI in Foundry Models) or custom-built using platforms like Azure Machine Learning. This layer also includes tools for fine-tuning, evaluating, and versioning models to ensure they meet performance and accuracy requirements. Microsoft Foundry, a unified Azure platform-as-a-service for enterprise AI operations, provides a comprehensive model catalog for application developers.

Compute Layer

AI applications require compute resources to train and run models. Microsoft provides several platform options:

  • Azure App Service for hosting web apps and APIs.
  • Azure Functions for serverless, event-driven execution of AI tasks.
  • Containers for scalable and portable deployment of AI models and services. Azure Container Instances (ACI) offers lightweight, serverless container execution, perfect for AI workloads needing rapid deployment and simple scaling. Azure Kubernetes Service (AKS) is a fully managed Kubernetes service that provides enterprise-level orchestration for AI workloads.

Note

Application programming interfaces (APIs) define the information that is required for one component to use the services of the other. APIs enable software components to communicate securely.

Integration & Orchestration Layer

The integration and orchestration layer connects models and data with business logic and user interfaces. Foundry plays a key role here by offering:

  • An agent Service for building intelligent agents that can reason and act.
  • AI Tools like speech, vision, and language APIs.
  • Software Development Kits (SDKs) and APIs for integrating AI capabilities into applications.
  • Portal tools for managing models, agents, and workflows.

By using Foundry to build their applications, developers can embed intelligence directly within the data layer for smarter, more responsive applications. Next, let's take a closer look at Foundry.