Build AI Apps with Azure Database for PostgreSQL

Modern AI applications require more than models. They need data platforms that can support both transactional workloads and intelligent features such as semantic search, vector embeddings, and grounded responses. This learning path explores how Azure Database for PostgreSQL can act as a foundation for building AI-powered applications.

We believe organisations that combine AI, cloud, and data effectively will move faster from insight to impact. In this learning path, learners will explore how to integrate PostgreSQL with Azure AI services, enabling applications that combine relational data with AI-driven capabilities. The focus is on practical developer scenarios, showing how to design scalable architectures that support retrieval, reasoning, and real-time application workflows.

Prerequisites

Participants should have:

• Basic application development experience
• Familiarity with relational databases and SQL concepts
• A general understanding of cloud platforms and artificial intelligence fundamentals
• Experience with APIs or backend development is beneficial but not required

Target audience

This learning path is designed for:

• Application developers building modern, data-driven applications
• Software engineers working with cloud-native architectures
• AI engineers integrating data platforms with AI services
• Technical professionals developing intelligent features using Azure

By the end of this learning path, learners will be able to:

• Describe how Azure Database for PostgreSQL supports AI application architectures
• Store, manage, and query vector embeddings within PostgreSQL
• Integrate PostgreSQL with Azure AI services and tools
• Build AI-powered features such as semantic search and retrieval
• Design retrieval-augmented generation patterns using relational and vector data
• Combine structured data with AI workflows to enable grounded responses

Introduction to AI application architectures on Azure

• Overview of modern AI application design patterns
• Role of data platforms in AI-powered applications
• Combining transactional systems with AI workloads
• Common use cases for intelligent, data-driven applications

Azure Database for PostgreSQL as an AI-ready data store

• Overview of Azure Database for PostgreSQL capabilities
• Supporting hybrid workloads with relational and vector data
• Managing structured and unstructured data for AI use cases
• Benefits of using managed open-source databases in Azure

Working with embeddings and vector data

• Introduction to vector embeddings and their role in AI
• Generating embeddings using Azure AI services
• Storing embeddings in PostgreSQL
• Performing similarity search and vector queries

Integrating PostgreSQL with Azure AI services

• Connecting database applications to Azure AI services
• Using APIs and SDKs for AI integration
• Combining database queries with AI model outputs
• Designing workflows that incorporate AI reasoning

Building AI-powered application patterns

• Implementing semantic search over application data
• Designing retrieval-augmented generation pipelines
• Grounding model responses using database content
• Enhancing user experiences with AI-driven features

Designing scalable and production-ready AI applications

• Architecting applications for scale and performance
• Managing data pipelines and query optimisation
• Monitoring and maintaining AI-enabled systems
• Applying best practices for security and reliability

Exams and assessments

There are no formal exams included in this learning path. Learners will complete knowledge checks and guided exercises to reinforce understanding of AI application patterns and PostgreSQL integration techniques.

Hands-on learning

This learning path includes:

• Practical exercises for working with embeddings and vector search
• Guided scenarios integrating PostgreSQL with Azure AI services
• Application-focused tasks for building AI-powered features
• Real-world examples of AI-enabled application architectures