Using AI Features in SQL Server 2025

Course Summary

Do you want to start working with AI-related features in SQL Server 2025?

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. It does not aim to turn SQL Server into an AI platform, nor does it attempt to teach model training, data science, or vendor-specific AI tooling.

This course is about using SQL Server 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 T-SQL 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?
— Have you been asked whether SQL Server 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 SQL Server 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 a SQL Server 2025 environment (trial or production) or a system you can install SQL Server on.

Greg Low
Instructor

Dr Greg Low is an international consultant, developer and trainer. He has been working in development since 1978, holds a PhD in Computer Science and a host of certifications. Greg is the founder and principal mentor for SQL Down Under, a Microsoft Data Platform MVP, host of the popular SQL Down Under and Fabric Down Under podcast series, and has worked with data for over three decades, and Fabric since before it was released.

Course Curriculum

Module 0: Getting started
4 Lessons
- Who is this course for?
  Preview
- Who is Greg?
  Preview
- What will I learn in this course?
  Preview
- Configuring your lab environment
  Start
Module 1: AI in the context of SQL Server
10 Lessons
- Module introduction
  Start
- AI features in SQL Server 2025
  Start
- What is new in SQL Server 2025
  Start
- What SQL Server is responsible for
  Start
- What SQL Server is not responsible for
  Start
- Misconceptions to avoid
  Start
- Positioning SQL Server correctly
  Start
- Lab 1
  Start
- Lab 1 video walkthroughs
  Start
- Quiz 1
  Start
Module 2: Vector data and embeddings fundamentals
15 Lessons
- Module introduction
  Start
- What is an AI model?
  Start
- What embeddings are and why they exist
  Start
- How vector similarity differs from relational comparison
  Start
- How SQL Server uses vector similarity
  Start
- Typical AI scenarios that involve vectors
  Start
- Why vectors work well for these scenarios
  Start
- When vectors are inappropriate
  Start
- How this applies to SQL Server
  Start
- How to think about vectors going forward
  Start
- Where embeddings models are hosted
  Start
- Common embeddings models
  Start
- Lab 2
  Start
- Lab 2 video walkthroughs
  Start
- Quiz 2
  Start
Module 3: Vector data types in SQL Server
17 Lessons
- Module introduction
  Start
- Vector data type basics
  Start
- Declaring VECTOR columns and variables
  Start
- How VECTOR values are represented
  Start
- Embeddings as derived data
  Start
- Practical table pattern for embeddings
  Start
- Vectors complement relational data
  Start
- Parallel tables and schema evolution
  Start
- Dimensionality is enforced
  Start
- Dimensionality changes are migration events
  Start
- Storage characteristics of VECTOR
  Start
- float16 vectors
  Start
- VECTOR limitations in SQL Server
  Start
- Lab 3
  Start
- Lab 3 solution
  Start
- Lab 3 video walkthroughs
  Start
- Quiz 3
  Start
Module 4: Querying vector data
16 Lessons
- Module introduction
  Start
- What similarity search means in T-SQL
  Start
- Exact similarity vs approximate search
  Start
- Writing similarity queries in T-SQL
  Start
- Using VECTOR_DISTANCE
  Start
- Combining vector similarity with relational predicates
  Start
- Using VECTOR_SEARCH
  Start
- Common query mistakes and inefficiencies
  Start
- How to reason about vector queries in T-SQL
  Start
- Using VECTOR_NORM
  Start
- Using VECTOR_NORMALIZE
  Start
- Using VECTORPROPERTY
  Start
- Lab 4
  Start
- Lab 4 solution
  Start
- Lab 4 video walkthroughs
  Start
- Quiz 4
  Start
Module 5: Vector indexing and performance
13 Lessons
- Module introduction
  Start
- Why vector indexing exists
  Start
- Vector index characteristics
  Start
- Operational constraints of vector indexes
  Start
- Creating a vector index
  Start
- Accuracy versus performance trade-offs
  Start
- CPU and memory impact of vector queries
  Start
- Monitoring and diagnosing vector query performance
  Start
- When vector indexes should not be used
  Start
- Lab 5
  Start
- Lab 5 solution
  Start
- Lab 5 video walkthroughs
  Start
- Quiz 5
  Start
Module 6: Generating embeddings outside SQL Server
9 Lessons
- Module introduction
  Start
- Why SQL Server does not generate embeddings
  Start
- Common embedding generation patterns
  Start
- Storing embeddings safely in SQL Server
  Start
- Refreshing and regenerating embeddings
  Start
- Handling changes in embedding models
  Start
- Versioning strategies for embeddings
  Start
- Re-embedding strategies at scale
  Start
- Quiz 6
  Start
Module 7: REST and HTTP basics
14 Lessons
- Module introduction
  Start
- What REST means in practice
  Start
- Core HTTP concepts
  Start
- JSON as a data format
  Start
- Authentication at a conceptual level
  Start
- Error handling and diagnostics
  Start
- How REST concepts apply to SQL Server integrations
  Start
- What SQL Server professionals need to know
  Start
- Using Caddy as a proxy
  Start
- Testing the Caddy proxy
  Start
- Lab 7
  Start
- Lab 7 solution
  Start
- Lab 7 video walkthroughs
  Start
- Quiz 7
  Start
Module 8: Calling AI services from SQL Server
19 Lessons
- Module introduction
  Start
- Synchronous vs asynchronous calls
  Start
- Patterns for invoking external AI services
  Start
- Security considerations for outbound calls
  Start
- Cost and throttling implications
  Start
- Operational risks and failure modes
  Start
- Designing SQL Server–safe AI integrations
  Start
- Using sp_invoke_external_rest_endpoint
  Start
- Calling sp_invoke_external_rest_endpoint
  Start
- Minimal structure of a REST call in T-SQL
  Start
- Retrieving text embeddings via SQL Server REST
  Start
- Using CREATE EXTERNAL MODEL
  Start
- Using AI_GENERATE_EMBEDDINGS
  Start
- Handling REST authentication
  Start
- Supported external model types
  Start
- Lab 8
  Start
- Lab 8 solution
  Start
- Lab 8 video walkthroughs
  Start
- Quiz 8
  Start
Module 9: Retrieval-augmented query patterns
12 Lessons
- Module introduction
  Start
- What RAG means in practical terms
  Start
- Designing SQL tables for retrieval
  Start
- Executing similarity search as part of a workflow
  Start
- Passing retrieved data to AI services
  Start
- Limitations of RAG for SQL Server
  Start
- Failure modes in SQL Server–based RAG
  Start
- How to use SQL Server effectively in RAG
  Start
- Lab 9
  Start
- Lab 9 solution
  Start
- Lab 9 video walkthroughs
  Start
- Quiz 9
  Start
Module 10: Security, governance, and operational concerns
9 Lessons
- Module introduction
  Start
- Protecting sensitive data used in AI workflows
  Start
- Controlling access to vector data
  Start
- Auditing AI-related queries
  Start
- Managing secrets and credentials
  Start
- Operational risks introduced by AI features
  Start
- Governance boundaries in AI-enabled systems
  Start
- How to approach AI governance with SQL Server
  Start
- Quiz 10
  Start
Module 11: When NOT to use AI features in SQL Server
8 Lessons
- Module introduction
  Start
- Why it matters to say no
  Start
- Scenarios where SQL Server is the wrong tool
  Start
- When a separate vector database makes more sense
  Start
- When Fabric or other platforms are more appropriate
  Start
- Cost, complexity, and maintenance trade-offs
  Start
- Decision checklist for architects and DBAs
  Start
- Quiz 11
  Start
Module 12: Using Data API Builder with SQL Server
23 Lessons
- Module introduction
  Start
- Data API Builder and SQL Server
  Start
- Database access through APIs
  Start
- What Data API Builder Offers
  Start
- Architecture using Data API Builder
  Start
- Installing Data API Builder
  Start
- Demonstration - Installing Data API Builder
  Start
- Creating a configuration file
  Start
- Using DAB CLI to create configuration
  Start
- Demonstration - Using DAB CLI to create configuration
  Start
- Common DAB CLI Commands
  Start
- Exposing SQL Server tables and views
  Start
- Exposing stored procedures
  Start
- Demonstration - Exposing database objects
  Start
- REST and GraphQL endpoints
  Start
- Endpoint querying and capabilities
  Start
- Demonstration - Querying the endpoints
  Start
- Authentication
  Start
- Security and permissions
  Start
- Data API Builder with SQL Server AI features
  Start
- Lab 12
  Start
- Lab 12 video walkthroughs
  Start
- Quiz 12
  Start
Module 13: Using SQL Server with AI Agents via MCP
18 Lessons
- Module introduction
  Start
- AI agents and external tools
  Start
- Why AI agents need external data
  Start
- Introducing Model Context Protocol
  Start
- MCP architecture
  Start
- SQL Server as an MCP tool source
  Start
- Reflection
  Start
- Local vs remote MCP servers
  Start
- MCP transport protocols
  Start
- Example agent workflow with SQL Server
  Start
- Security considerations
  Start
- Installing a SQL Server MCP Server
  Start
- SQL MCP Server with Data API Builder
  Start
- Demonstration - Checking and testing MCP server
  Start
- Demonstration - Using an LLM with the MCP server data
  Start
- Lab 13
  Start
- Lab 13 video walkthroughs
  Start
- Quiz 13
  Start
Module 14: Next steps
1 Lesson
- Summary and further steps
  Start

Course Pricing

Most popular

AI Features Course Access

$195 USD