Which vector databases does this skill support?

This skill provides specialized implementation patterns and templates for Pinecone, Qdrant, and PostgreSQL using the pgvector extension.

How does it help with scaling to millions of vectors?

It provides guidance on choosing the right index types, such as HNSW for speed or IVF+PQ for memory efficiency, allowing your search to remain fast as data grows.

Does it support hybrid search?

Yes, the skill includes patterns for hybrid search that combine dense vector embeddings with sparse keyword-based search for better retrieval performance.

Can I use this for RAG applications?

Yes, it is specifically designed for RAG, including patterns for initial retrieval, over-fetching, and reranking with cross-encoders to improve LLM context quality.

Similarity Search Patterns

Name: Similarity Search Patterns
Author: nguyendinhquocx

bynguyendinhquocx

0•

Ciencia de Datos y ML

Implements high-performance vector similarity search and RAG retrieval patterns using popular vector databases and indexing strategies.

This skill provides standardized implementation patterns for vector similarity search, which is a foundational component for building modern AI applications like Retrieval-Augmented Generation (RAG) systems and recommendation engines. It includes production-ready templates for leading vector stores such as Pinecone, Qdrant, and pgvector. By following these patterns, developers can easily manage distance metrics, select appropriate index types like HNSW or IVF, and implement advanced techniques such as hybrid search and cross-encoder reranking to ensure high precision and low latency in production environments.

Características Principales

01Support for advanced RAG retrieval and reranking patterns

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

03Index optimization strategies using HNSW and scalar quantization

040 GitHub stars

05Hybrid search patterns combining semantic vectors with keyword filtering

06Expert guidance on selecting optimal distance metrics (Cosine, L2, Dot Product)

Casos de Uso

01Implementing high-accuracy RAG workflows for LLM-powered applications

02Developing real-time recommendation engines based on embedding similarities

03Building semantic search systems for large-scale enterprise documentation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nguyendinhquocx/code-ai similarity-search-patterns

For use in Claude.ai and ChatGPT

Download Skill