Author: Microsoft Developer - Bewertung: 6x - Views:162
Host Justin Garrett sits down with Microsoft MVP Andrew Pruski to explore the power of vector search in SQL Server 2025 and how it’s changing the way we find meaning in data. In this episode, Andrew walks through a fun but practical demo called the “Burrito Bot,” a semantic search application that uses embeddings and vector indexing to recommend restaurants based on concepts like “cozy atmosphere” or “romantic date spot.” Along the way, you’ll learn how vector search works, how embeddings capture meaning, and how to scale efficiently using SQL Server’s new capabilities. Whether you’re a DBA, developer, or AI enthusiast, this episode breaks down complex concepts like high-dimensional vectors, cosine similarity, and approximate nearest neighbor search into approachable, real-world examples.
In this episode, you’ll learn:
✅How vector search enables semantic (meaning-based) search beyond traditional keyword matching
✅How to generate and use embeddings in SQL Server 2025 for real-world applications
✅How cosine similarity and vector distance power more accurate search results
✅Best practices for scaling with vector indexes and approximate nearest neighbor (ANN) search
✅How to build a practical AI app using RAG (Retrieval-Augmented Generation) with Azure AI Foundry
✅Why choosing the right model for the job (cost vs performance) matters in AI solutions
🔗 Chapter Markers:
00:07 - Intro: Meet Andrew Pruski + AI Breakthrough Moment
01:09 - Vector Search vs Keyword Search (Why It Matters)
02:10 - Live Demo: Building the Burrito Bot App
05:21 - Data + Embeddings: How AI Understands Meaning
10:39 - Visualizing Vectors and Similarity (Made Simple)
18:07 - Scaling Search with Vector Indexes and ANN
24:11 - RAG + GPT: Smarter Search Experiences
33:33 - Future of AI Search + MVP Journey
👉 Subscribe for more MVP insights and AI-powered development tips!
#VectorSearch #SQLServer2025 #SemanticSearch #VectorSearchDemo #SemanticSearchAI #EmbeddingsExplained #AzureAIFoundry #AzureOpenAIService #RAGArchitecture #AIInDatabases #SQLServerAI #ApproximateNearestNeighbor #CosineSimilarity #MachineLearningDemo #AIForDevelopers #DataPlatform #MicrosoftAzure #CloudAI #DatabaseAI #AIEngineering #DevRel #TechDeepDive
🔗 Resources & Links
🎁 Free Microsoft Foundry Trial: https://aka.ms/devrelft
📊Visualization app: projector.tensorflow.org
✏️Andrew's repo: github.com/dbafromthecold/aipoweredsearch
🧾Vector search docs: learn.microsoft.com/sql/sql-server/ai/vectors?view=sql-server-ver17#vector-search
About Andrew
Andrew is a Microsoft Data Platform MVP and Docker Captain who helps organise Data Ceili and EightKB. He is interested in all things database/kubernetes/container related and shares his passion for these topics by speaking at events across the world. Originally from Wales but now exploring Ireland.
Bluesky: @dbafromthecold.com
Blog: dbafromthecold.com
About Justin
Justin Garrett is host of MVP Unplugged, Principal PM in Developer Relations, in Microsoft Cloud + AI. Justin’s career at Microsoft spans 20 years across Windows, Bing, Edge, Web Platform, Academic / University Relations, Cloud Advocacy, and most recently a leader of the MVP Program. In his spare time, he enjoys volunteering in STEM, coaching youth soccer, planting trees and is an avid mountaineer and outdoorsy person.
🔗 Chapter Markers:
00:07 - Intro: Meet Andrew Pruski + AI Breakthrough Moment
01:09 - Vector Search vs Keyword Search (Why It Matters)
02:10 - Live Demo: Building the Burrito Bot App
05:21 - Data + Embeddings: How AI Understands Meaning
10:39 - Visualizing Vectors and Similarity (Made Simple)
18:07 - Scaling Search with Vector Indexes and ANN
24:11 - RAG + GPT: Smarter Search Experiences
33:33 - Future of AI Search + MVP Journey
🔗 Resources & Links
🎁 Free Microsoft Foundry Trial: https://aka.ms/devrelft
📊Visualization app: projector.tensorflow.org
✏️Andrew's repo: github.com/dbafromthecold/aipoweredsearch
🧾Vector search docs: learn.microsoft.com/sql/sql-server/ai/vectors?view=sql-server-ver17#vector-search
Bluesky: @dbafromthecold.com
Blog: dbafromthecold.com
SOCIAL SHARE CARD GENERATOR