(TL;DR; Video trailer for the new course is here.)
We’re building on our library of new SQL Server 2025 and Azure SQL Database with a more in-depth course on Vector Search. On Monday, Joe’s next course, AIVSP: Vector Search in Practice, will be available!
This online course builds on AIVSE: Vector Search Essentials to cover production patterns for vector search in SQL Server 2025. Designed for SQL Server DBAs and developers who have completed the prerequisite course, you’ll learn how to select embedding models, design schemas, measure search quality, handle edge cases, and implement advanced patterns like reranking and RAG. Through detailed explanations and hands-on demonstrations, you’ll gain the practical knowledge needed to deploy and maintain vector search in production environments.
The hands-on lab database and all demo scripts are available on GitHub. Whether you’re running SQL Server locally, on a VM, or in Azure, the core concepts and query patterns you’ll learn apply across all deployment options.
You’ll learn the following:
- Embedding model selection and evaluation
- Schema design patterns for long-term maintainability
- Measuring search quality with precision, recall, MRR, and nDCG metrics
- Diagnosing vector search failure modes
- Solving the post-filtering problem with candidate pool sizing
- Reranking strategies including Reciprocal Rank Fusion
- Document chunking approaches
- RAG pipeline implementation with local LLMs
- Embedding freshness maintenance
- Vector database selection criteria
A video trailer and more information about the course is here.
It’ll be available from Monday, February 16th in our shop, for 1-year or lifetime access, and in a bundle with the AIVSE course. If you’re a Blackbelt Lifetime subscriber, login and use the discount code ‘bbai’ to save $50 on the price (it automatically checks your account email), otherwise, check the newsletter on Tuesday morning for a $50 discount coupon.
Any questions, let me know.
Enjoy!