vector-db

Select and optimize vector databases for approximate nearest neighbor tasks.

Updated Jun 5, 2026

Search Tags:#vector database #performance tuning #data indexing #ANN #approximate nearest neighbor

Ready to run

npx skills add https://github.com/hung-phan/system-skills --skill vector-db-hung-phan

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: vector-db
Source: https://github.com/hung-phan/system-skills/tree/main/skills/system-review/references/data-systems/vector-db
Command: npx skills add https://github.com/hung-phan/system-skills --skill vector-db-hung-phan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps users choose the right vector database and optimize its performance for approximate nearest neighbor (ANN) tasks, addressing issues like high latency, low recall, and data sharding.

Core Features & Use Cases

Vector Database Selection: Provides guidance on selecting the appropriate vector database based on vector count, data characteristics, and performance requirements.
Performance Tuning: Offers best practices for tuning index parameters and query strategies to achieve optimal performance.
Use Case: When dealing with a dataset of millions of embeddings, this Skill helps determine whether to use pgvector, Pinecone, Qdrant, or Milvus, and how to configure it for the best results.

Quick Start

Analyze the performance of your vector database using the vector-db skill and adjust the index parameters accordingly.

Frequently Asked Questions about vector-db

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

FAQPage Schema

How do I choose the right vector database for approximate nearest neighbor tasks?▼

To choose a vector database for approximate nearest neighbor tasks, evaluate your vector count, data characteristics, and performance requirements. This Skill guides selection among options like pgvector, Pinecone, Qdrant, and Milvus based on your specific recall and latency needs.

What's the best way to tune vector database index parameters for high recall and low latency?▼

Tune vector database index parameters by applying best practices for query strategies and indexing configurations. This Skill analyzes your setup and helps adjust parameters to achieve optimal performance, balancing high recall with low latency and efficient resource usage.

How do I optimize vector database performance for millions of embeddings?▼

Optimize vector database performance for millions of embeddings by configuring proper indexing and query strategies. This Skill helps determine whether to use pgvector, Pinecone, Qdrant, or Milvus, and how to tune them for the best results at scale.

Does this approach support hybrid search across different data types?▼

Yes, optimizing vector databases for approximate nearest neighbor tasks includes support for hybrid search. This Skill handles data of various sizes and types, focusing on configuring indexing and query strategies to maintain performance across diverse datasets.

Why does my vector database have low recall and high latency during data indexing?▼

Low recall and high latency during vector database indexing often result from unoptimized index parameters and query strategies. This Skill provides performance tuning best practices to address these issues, ensuring efficient data sharding and resource usage.