Similarity Search Patterns

Solid patterns for building semantic search systems with vector databases like Pinecone and Qdrant. Covers the essentials you actually need: distance metrics, indexing strategies, batch operations, and hybrid search with reranking. The templates handle real production concerns like quantization for memory efficiency, filtering, and proper batching. One thing I appreciate is it doesn't oversell approximate search accuracy, giving you realistic recall expectations for different index types. If you're building RAG systems or recommendation engines, this gives you working code that scales beyond toy datasets.

skills.sh installs: 11K
GitHub stars: 42.4K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add wshobson/agents --skill similarity-search-patterns --agent claude-code

Installs into .claude/skills of the current project.

Files

Similarity Search Patterns

Patterns for implementing efficient similarity search in production systems.

When to Use This Skill

Building semantic search systems
Implementing RAG retrieval
Creating recommendation engines
Optimizing search latency
Scaling to millions of vectors
Combining semantic and keyword search

Core Concepts

1. Distance Metrics

| Metric | Formula | Best For | | ------------------ | ------------------ | --------------------- | --- | -------------- | | Cosine | 1 - (A·B)/(‖A‖‖B‖) | Normalized embeddings | | Euclidean (L2) | √Σ(a-b)² | Raw embeddings | | Dot Product | A·B | Magnitude matters | | Manhattan (L1) | Σ | a-b | | Sparse vectors |