databricks-vector-search

Create and query Databricks Vector Search indexes for RAG applications.

Updated Mar 12, 2026

Search Tags:#semantic search #embeddings #rag #vector search #databricks #data indexing

Ready to run

npx skills add https://github.com/juanlamadrid20/coda --skill databricks-vector-search-juanlamadrid20

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: databricks-vector-search
Source: https://github.com/juanlamadrid20/coda/tree/main/.claude/skills/databricks-vector-search
Command: npx skills add https://github.com/juanlamadrid20/coda --skill databricks-vector-search-juanlamadrid20

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill simplifies the creation and querying of vector search indexes within Databricks, enabling efficient semantic search and RAG applications.

Core Features & Use Cases

Vector Index Management: Create and manage both Standard and Storage-Optimized endpoints and indexes.
Data Synchronization: Supports Delta Sync for automatic updates from Delta tables and Direct Access for manual control.
Querying: Perform semantic search using text, query vectors, or hybrid approaches with advanced filtering.
Use Case: You need to build a question-answering system over your company's documentation. This Skill allows you to create a vector index from your Delta tables, automatically embed the text, and then query it with natural language questions to retrieve relevant information.

Quick Start

Use the databricks-vector-search skill to create a standard endpoint named 'my-vs-endpoint'.

Frequently Asked Questions about databricks-vector-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I create a vector search index in Databricks for RAG applications?▼

To create a vector search index for RAG applications in Databricks, configure either a Delta Sync or Direct Access index under a Standard or Storage-Optimized endpoint. This syncs Delta tables and manages embeddings to enable efficient similarity matching.

Can I automatically sync my Delta table embeddings to a Databricks vector index?▼

Yes, you can automatically sync Delta table embeddings using the Delta Sync index type with managed embeddings. This ensures your Databricks vector search index automatically updates whenever the underlying Delta table data changes.

What is the difference between Standard and Storage-Optimized vector search endpoints?▼

Standard and Storage-Optimized endpoints dictate the infrastructure scaling and latency characteristics for your Databricks vector search. Storage-Optimized endpoints provide cost-efficient scaling for larger vector indexes compared to Standard endpoints.

Does Databricks vector search support hybrid search and advanced filtering?▼

Yes, Databricks vector search supports hybrid search and advanced filtering. You can perform semantic search using text, query vectors, or hybrid approaches alongside filters to retrieve highly relevant similarity matches from your index.

When should I use Direct Access instead of Delta Sync for data indexing in Databricks?▼

Use Direct Access for manual control over data indexing when you need to push vectors directly without Delta table synchronization. Choose Delta Sync when you want your Databricks vector search index to automatically reflect source table updates.