skills.rest
spideynolove
spideynoloveCOMMUNITY
Explore all skills by this author →
Website𝕏@spideynoloveGitHub

similarity-search-patterns

Implement vector similarity search with distance metrics and index templates.

2|1|Updated Feb 27, 2026
Search Tags:#indexing#semantic-search#vector-databases#retrieval-augmented-generation#similarity-search#nearest-neighbor
One-click install
npx skills add https://github.com/spideynolove/claude-code-in-action --skill similarity-search-patterns-spideynolove
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: similarity-search-patterns
Source: https://github.com/spideynolove/claude-code-in-action/tree/main/27-tdd-conductor-llmdev/.claude/skills/similarity-search-patterns
Command: npx skills add https://github.com/spideynolove/claude-code-in-action --skill similarity-search-patterns-spideynolove

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Efficiently implement, optimize, and operate vector-based similarity search across large data collections.

Core Features & Use Cases

  • Flexible distance metrics (cosine, euclidean, dot product, manhattan) for embedding comparisons.
  • Diverse index types (Flat, HNSW, IVF+PQ) with guidance on when to use each.
  • Practical templates and patterns for common stacks ( Pinecone, Qdrant, pgvector, Weaviate ) to accelerate production deployment.
  • Use cases include semantic search, retrieval-augmented generation, and large-scale recommendations.

Quick Start

Configure a vector store and run a sample search using the provided templates.

Frequently Asked Questions about similarity-search-patterns

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

FAQPage Schema
How do I implement similarity search in a vector database for RAG pipelines?▼

To implement similarity search for retrieval-augmented generation, configure a vector store using provided templates for Pinecone, Qdrant, pgvector, or Weaviate, then run queries with appropriate distance metrics like cosine or dot product.

When should I use HNSW vs IVF+PQ indexes for large-scale vector search?▼

Use HNSW indexes for fast approximate nearest neighbor searches with high recall, and switch to IVF+PQ when managing millions of vectors to optimize memory footprint and storage efficiency in your vector database.

Can I use pgvector for semantic search across millions of vectors?▼

Yes, pgvector supports semantic search across large data collections. You can apply practical templates to configure distance metrics and index types within PostgreSQL to scale your embedding comparisons effectively.

What distance metrics work best for embedding comparisons in vector search?▼

Cosine similarity, euclidean distance, dot product, and manhattan distance are supported. Choose cosine for normalized text embeddings, dot product for maximum inner product search, and euclidean for general spatial distance calculations.

How do I optimize vector database retrieval for recommendation engines?▼

Optimize recommendation engine retrieval by selecting efficient index types like HNSW or IVF+PQ and applying the correct distance metrics to compare user and item embeddings quickly across millions of vectors.

Related Skills

Explore More Software Engineering Skills→
lewisperez999lewisperez999

vector-embeddings

Generate text embeddings with OpenAI and search by cosine similarity in Upstash Vector or pg-vector.

Community
Intermediate
anukkrit149anukkrit149

rag-implementation

Build Retrieval-Augmented Generation systems grounding LLM answers in vector databases.

Community
Advanced
tranhieutttranhieutt
70

vector-database-engineer

Implement vector database workflows for semantic search retrieval.

Community
Advanced
winsorllcwinsorllc

Embedding Search

Provides semantic search across documents with Tunrsity and Tunrsity-like features.

Community
Intermediate