Using AI Features in Azure HorizonDB

Store, retrieve, and integrate data for AI systems using Azure HorizonDB

Enroll now for $195 USD !

Course Summary

Do you want to start working with AI-related features in Azure HorizonDB? Or is someone in your organization likely to start asking you about using databases this way? Either way, you need to be prepared!

As these features are relatively new, the focus of this course is on design and architecture decisions that remain valid, even as individual features, syntax, and external AI services evolve. But we’ll show you how to implement things right now.

This course is about using Azure HorizonDB to store, retrieve, and integrate data for AI systems, in a way that is reliable, understandable, and operationally sound. We assume you already understand SQL Server fundamentals and are comfortable working with PostgreSQL and database design.

  • Are you responsible for deciding where AI-related logic should live, and where it should not?
  • Do you need to support vector search, retrieval, or AI integration without turning the database into an experimental system?
  • Do you want to understand Data API Builder and configuring an Azure HorizonDB MCP server?
  • Have you been asked whether databases should be used for embeddings, RAG patterns, or AI orchestration, and wanted a clear answer that you can explain?

If that matches the kind of responsibility you have, you’re in the right place. You’ll get clear explanations, hands-on exercises, and short quizzes to make sure you not only understand the concepts but can confidently apply them in real Azure HorizonDB environments.

And while the practical exercises are optional, we strongly encourage completing them. They are designed to be straightforward and focused on realistic scenarios, rather than tooling complexity. Setup is easy. The main requirement is access to an Azure subscription where you can deploy an Azure HorizonDB resource.


Enroll now for $195 USD !

Modules and Lessons

Module 0: Getting Started

Who is this course for ?
Who is Greg ?
What will I learn in this course ?
Configuring your lab environment

Module 1: AI in the context of Azure HorizonDB

Module introduction
AI features in Azure HorizonDB
What the database is responsible for
What the database is not responsible for
Misconceptions to avoid
Positioning Azure HorizonDB correctly
Lab 1 video walkthroughs
Lab 1

  • Exercise 0: Overview
  • Exercise 1: Provisioning HorizonDB
  • Exercise 2: Enabling the required extensions

Quiz 1

Module 2: Vector data and embeddings fundamentals

Module introduction
What is an AI model ?
What embeddings are and why they exist
How vector similarity differs from relational comparison
Typical AI scenarios that involve vectors
When vectors are inappropriate
Where embeddings models are hosted
Common embeddings models
Lab 2 video walkthroughs
Lab 2

  • Exercise 0: Overview
  • Exercise 1: Generating and inspecting raw embeddings

Quiz 2

Module 3: Vector data types in PostgreSQL

Module introduction
Vector data type basics
pgvector extension basics
Declaring vector columns and variables in PL/pgSQL
How vector values are represented
Practical table pattern for embeddings
Dimensionality is enforced
Dimensionality changes are migration events
Storage characteristics and limitations of pgvector
Lab 3 video walkthroughs
Lab 3

  • Exercise 0: Overview
  • Exercise 1: Creating tables with vector data types

Quiz 3

Module 4: Querying vector data

Module introduction
What similarity search means in PostgreSQL
Writing similarity queries in SQL
Exact similarlity (KNN) vs approximate search (ANN)
Combining vector similarity with relational predicates
Common query mistakes and inefficiencies
Lab 4 video walkthroughs
Lab 4

  • Exercise 0: Overview
  • Exercise 1: Populating vector columns and executing distance queries

Quiz 4

Module 5: Vector indexing and performance

Module introduction
Why vector indexing exists
Traditional PostgreSQL vector indexes - HNSW and IVFFlat
Introduction to DiskANN in HorizonDB
Creating a DiskANN index
Accuracy versus performance trade-offs
CPU and memory impact of vector queries
Monitoring and diagnosing vector query performance
Lab 5 video walkthroughs
Lab 5

  • Exercise 0: Overview
  • Exercise 1: Implementing DiskANN and comparing execution plans

Quiz 5

Module 6: Integrating External AI Services

Module introduction
Why HorizonDB abstracts REST API complexity
Understanding the azure_ai schema
Security and managed identities
Cost and throttling implications of outbound calls
Operational risks and failure modes
Lab 6 video walkthroughs
Lab 6

  • Exercise 0: Overview
  • Exercise 1: Configuring Azure AI Foundry credentials via managed identity

Quiz 6

Module 7: Calling AI services from HorizonDB

Module introduction
Synchronous vs asynchronous calls in PostgreSQL
Retrieving text embeddings via azure_openai.create_embeddings
Invoking chat models directly from SQL
Handling changes in embedding models and versioning strategies
Re-embedding strategies at scale
Lab 7 video walkthroughs
Lab 7

  • Exercise 0: Overview
  • Exercise 1: Populating vector columns natively using azure_openai

Quiz 7

Module 8: Durable AI pipelines

Module introduction
Challenges with external ETL for AI workflows
pg_durable execution engine
Defining a pipeline with sources, steps, and sinks
Handling automatic chunking and embedding in SQL
Managing pipeline state and recovery
Monitoring and troubleshooting persistent workflows
Lab 8 video walkthroughs
Lab 8

  • Exercise 0: Overview
  • Exercise 1: Building a continuous chunk-and-embed pipeline for incoming documents

Quiz 8

Module 9: Retrieval-augmented query patterns

Module introduction
What RAG means in practical terms
Executing similarity search in a HorizonDB workflow
In-database re-ranking
Passing retrieved data back to the app tier or to LLMs
Lab 9 video walkthroughs
Lab 9

  • Exercise 0: Overview
  • Exercise 1: Executing an end-to-end RAG query using indexed vectors

Quiz 9

Module 10: Security, governance, and operational concerns

Module introduction
Protecting sensitive data used in AI workflows
Applying PostgreSQL RLS to vector searches
Auditing AI-related queries
Governance boundaries in AI-enabled systems
Quiz 10

Module 11: When NOT to use AI features in HorizonDB

Module introduction
Why it matters to say no
Scenarios where HorizonDB is the wrong tool
When a specialized vector database makes more sense
Cost, complexity, and mainteance trade-offs
Decision checklist for architects and DBAs
Quiz 11

Module 12: Using Data API Builder with HorizonDB

Module introduction
Data API Builder and PostgreSQL
Architecture and configuration for HorzionDB
Exposing tables, views, and functions via REST and GraphQL
Authentication and security
Lab 12 video walkthroughs
Lab 12

  • Exercise 0: Overview
  • Exercise 1: Exposing HorizonDB vector search via a GraphQL endpoint

Quiz 12

Module 13: Using HorizonDB with AI Agents via MCP

Module introduction
AI agents and external tools
SQL Server MCP vs PostgreSQL MCP implementation
MCP architecture for HorizonDB
Example agent workflow querying pgvector data
Lab 13 video walkthroughs
Lab 13

  • Exercise 0: Overview
  • Exercise 1: Configuring an MCP server to expose HorizonDB search to a local LLM

Quiz 13

Module 14: Next steps

Summary and further steps


Enroll now for $195 USD !