Can this skill help with RAG performance optimization?

Yes, it provides optimized indexing strategies like HNSW and quantization techniques to improve retrieval speed and memory efficiency in RAG systems.

What vector databases are supported by this skill?

The skill includes detailed implementation templates and best practices for Pinecone, Qdrant, and PostgreSQL with the pgvector extension.

How do I choose the right distance metric?

The skill includes a reference guide comparing Cosine, Euclidean (L2), Dot Product, and Manhattan metrics to help you match the right metric to your specific embedding model.

Does it support hybrid search patterns?

Yes, the skill covers patterns for combining semantic (dense) search with metadata filtering and sparse vector search for more accurate results.

Vector Similarity Search Patterns

Name: Vector Similarity Search Patterns
Author: ccf

byccf

0•

데이터 과학 및 ML

Implements high-performance similarity search and vector database patterns for RAG and semantic retrieval systems.

This skill provides specialized implementation patterns for vector similarity search, covering essential distance metrics and indexing strategies like HNSW and IVF+PQ. It offers production-ready templates for industry-leading vector stores including Pinecone, Qdrant, and pgvector, enabling developers to build scalable Retrieval-Augmented Generation (RAG) systems, recommendation engines, and semantic search applications with optimized performance. Whether you are scaling to millions of vectors or optimizing search latency, these patterns provide the domain-specific guidance needed to integrate robust vector search capabilities into any LLM-powered application.

주요 기능

010 GitHub stars

02Production-ready templates for Pinecone, Qdrant, and pgvector

03Optimized indexing patterns including HNSW and Quantized (IVF+PQ)

04Advanced search techniques like cross-encoder reranking

05Hybrid search implementation for combining semantic and keyword results

06Comprehensive distance metric guidance (Cosine, L2, Dot Product)

사용 사례

01Developing AI-powered recommendation engines and content discovery

02Building Retrieval-Augmented Generation (RAG) retrieval pipelines

03Implementing semantic search for large-scale document repositories

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ccf/claude-code-ccf-marketplace similarity-search-patterns

For use in Claude.ai and ChatGPT

Download Skill