Vector Database Master

概要

This skill provides expert guidance on selecting, implementing, and optimizing vector databases like Chroma, Pinecone, and Weaviate for AI-driven applications. It covers essential indexing strategies such as HNSW and IVF, distance metrics for semantic similarity, and advanced retrieval techniques like hybrid search and metadata filtering. Designed for AI engineers, it ensures that your Retrieval-Augmented Generation (RAG) systems are scalable, efficient, and capable of high-precision data retrieval in production environments.

主な機能

• Indexing strategy optimization for HNSW, IVF, and Flat indexes
• Advanced metadata filtering and batch processing techniques
• Implementation patterns for Chroma, Pinecone, Weaviate, and Milvus
• Hybrid search implementation combining dense and sparse retrieval
• 2 GitHub stars
• Semantic search tuning using Cosine, L2, and Dot Product metrics

ユースケース

• Implementing semantic similarity search for large-scale document repositories
• Building production-grade Retrieval-Augmented Generation (RAG) systems
• Optimizing vector database performance and latency for real-time AI agents