skills.rest
leonardoteodoroo
leonardoteodorooCOMMUNITY
Explore all skills by this author →
GitHub

similarity-search-patterns

Implement similarity search patterns for vector embeddings with distance metrics and index templates.

Updated Feb 3, 2026
Search Tags:#indexing#semantic-search#vector-databases#similarity-search#nearest-neighbor#recommendation-systems
One-click install
npx skills add https://github.com/leonardoteodoroo/amino-advanced --skill similarity-search-patterns-leonardoteodoroo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: similarity-search-patterns
Source: https://github.com/leonardoteodoroo/amino-advanced/tree/main/.agent/skills/similarity-search-patterns
Command: npx skills add https://github.com/leonardoteodoroo/amino-advanced --skill similarity-search-patterns-leonardoteodoroo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Patterns for implementing efficient similarity search in production systems.

Core Features & Use Cases

  • Distance Metrics: Cosine, Euclidean, Dot Product, and Manhattan.
  • Index Types: Flat, HNSW, IVF+PQ with guidance on performance and recall.
  • Templates: Example implementations for Pinecone, Qdrant, pgvector, and Weaviate.

Use cases include semantic search, RAG retrieval, recommendations, and large-scale vector indexing, as well as strategies to combine semantic and keyword search.

Quick Start

Run a quick demo that performs a nearest-neighbor search on a small vector set.

Frequently Asked Questions about similarity-search-patterns

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

FAQPage Schema
How do I choose the right distance metric for vector similarity search?▼

Vector similarity search relies on distance metrics to measure embedding similarity. This skill provides guidance on selecting and implementing Cosine, Euclidean, Dot Product, and Manhattan metrics based on your specific data and retrieval requirements.

What is the best way to scale nearest-neighbor search for large-scale vector indexing?▼

Scaling nearest-neighbor search requires efficient index types to balance performance and recall. This skill provides patterns for implementing Flat, HNSW, and IVF+PQ indexes to handle large-scale vector embeddings in production systems.

Does this skill provide implementation templates for Pinecone, Qdrant, pgvector, and Weaviate?▼

Yes, this skill provides ready-to-use implementation templates for Pinecone, Qdrant, pgvector, and Weaviate. These templates help configure vector databases for semantic search, RAG retrieval, and recommendation systems.

How do I combine semantic search with keyword search for better retrieval results?▼

Combining semantic and keyword search improves retrieval accuracy by blending vector similarity with exact term matching. This skill offers strategies to integrate both approaches within your vector database workflows.

When should I use HNSW over IVF+PQ for vector database indexing?▼

Choosing between HNSW and IVF+PQ depends on your specific performance and recall trade-offs. This skill provides guidance on selecting the appropriate index type based on your production scale and latency requirements.

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
jason660519jason660519

faiss

Train FAISS indexes and run nearest-neighbor search over large embedding datasets.

Community
Advanced
ccfccf

similarity-search-patterns

Detail distance metrics and index types for vector similarity search across Pinecone, Qdrant, pgvector, and Weaviate.

Community
Advanced
Division6066Division6066

hybrid-search-implementation

Combine vector similarity with keyword matching for hybrid search retrieval.

Community
Advanced