skills.rest
lucaslessachaves
lucaslessachavesCOMMUNITY
Explore all skills by this author →
GitHub

databricks-vector-search

Create and query Databricks vector search indexes with Delta Sync and Direct Access.

1|Updated Dec 15, 2025
Search Tags:#rag#endpoints#embedding#semantic-search#vector-search#databricks#delta-lake
One-click install
npx skills add https://github.com/lucaslessachaves/default --skill databricks-vector-search-lucaslessachaves
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: databricks-vector-search
Source: https://github.com/lucaslessachaves/default/tree/main/.claude/skills/databricks-vector-search
Command: npx skills add https://github.com/lucaslessachaves/default --skill databricks-vector-search-lucaslessachaves

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Databricks Vector Search enables teams to pattern, provision, and query scalable vector indexes for RAG and semantic search workloads, reducing complexity and time-to-value.

Core Features & Use Cases

  • Create compute endpoints (STANDARD or STORAGE_OPTIMIZED) and index types (DELTA_SYNC, DIRECT_ACCESS) to suit latency and scale requirements.
  • Manage embeddings through Delta Sync (managed or self-managed) or direct vector storage for flexible architectures, with hybrid search support.
  • End-to-end workflows for source data ingestion, embedding generation, index synchronization, and query-time retrieval in production-ready pipelines.

Quick Start

Run through a quick demo by provisioning a STANDARD endpoint, creating a DELTA_SYNC index with managed embeddings, and issuing your first query.

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 a RAG application?▼

To create a vector search index in Databricks for RAG, you need a Delta table as your source, an embedding model, and the SDK tooling to provision an endpoint and build either a DELTA_SYNC or DIRECT_ACCESS index.

What is the difference between Delta Sync and Direct Access vector indexes?▼

Delta Sync indexes automatically synchronize embeddings from a Delta table, whereas Direct Access indexes require pre-computed embeddings for direct vector storage, offering flexible architecture choices for semantic search workloads.

When should I use STORAGE_OPTIMIZED versus STANDARD endpoints for vector search?▼

Use STANDARD endpoints for general vector search latency requirements, and choose STORAGE_OPTIMIZED endpoints when your semantic search workload demands specific scale and storage characteristics tailored to large vector datasets.

Can I use self-managed embeddings with Databricks Vector Search?▼

Yes, Databricks Vector Search supports self-managed embeddings through Delta Sync, or you can supply pre-computed embeddings directly to a DIRECT_ACCESS index, bypassing managed embedding generation entirely.

How does query-time filtering work for semantic search in Databricks?▼

Query-time filtering in Databricks Vector Search allows you to apply constraints during retrieval across both storage-optimized and standard endpoints, refining semantic search results dynamically within your RAG pipelines.

Related Skills

Explore More Data & Analytics Skills→
RamVegirajuRamVegiraju
4

databricks-vector-search

Create and query Databricks vector search endpoints and indexes.

Community
Advanced
giuseppe-trisciuogliogiuseppe-trisciuoglio
322

qdrant-vector-database-integration

Integrate Qdrant vector database with Java Spring Boot and LangChain4j for embedding storage and similarity search.

Community
Advanced
datasciencemonkeydatasciencemonkey
27

databricks-lakebase-provisioned

Provision Lakebase PostgreSQL instances with OAuth and Unity Catalog integration.

Community
Advanced
pacphipacphi
6

AgentDB Vector Search

Implement semantic vector search with HNSW indexing in Node.js.

Community
Advanced