vector-database-engineer

Design and optimize vector databases for semantic search and RAG applications.

Updated Mar 17, 2026

Search Tags:#indexing #embeddings #RAG #semantic-search #vector-databases #hybrid-search

Ready to run

npx skills add https://github.com/HemantSudarshan/Dhumichatbot --skill vector-database-engineer-hemantsudarshan

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: vector-database-engineer
Source: https://github.com/HemantSudarshan/Dhumichatbot/tree/main/skills/02-data/vector-database-engineer
Command: npx skills add https://github.com/HemantSudarshan/Dhumichatbot --skill vector-database-engineer-hemantsudarshan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Many organizations struggle to scale semantic search and RAG without optimized vector databases, embeddings, and indexing strategies.

Core Features & Use Cases

Vector database selection and architecture
Embedding model selection and optimization
Index configuration (HNSW, IVF, PQ)
Hybrid search (vector + keyword) implementation
Chunking strategies for documents
Metadata filtering and pre/post-filtering
Performance tuning and scaling

Quick Start

Set up a baseline vector store and begin embedding 10k documents to evaluate indexing strategies.

Frequently Asked Questions about vector-database-engineer

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

FAQPage Schema

How do I optimize a vector database for RAG applications?▼

Optimize vector databases for RAG by selecting appropriate embedding models, configuring HNSW or IVF indexes, and implementing document chunking. This ensures fast semantic search retrieval and scalable architecture across platforms like Pinecone, Weaviate, and Qdrant.

What is the best way to implement hybrid search with vector databases?▼

Implement hybrid search by combining vector similarity with keyword filtering. You achieve this by applying metadata pre-filtering or post-filtering techniques alongside vector embeddings to refine search accuracy and restrict results within your vector database.

How does HNSW index configuration affect semantic search performance?▼

HNSW index configuration impacts semantic search latency and accuracy by structuring graph layers for rapid vector traversal. Tuning HNSW parameters alongside embedding strategies ensures efficient retrieval and scalable performance as your document volume grows.

Can I use pgvector for large-scale RAG applications?▼

Yes, pgvector supports large-scale RAG applications when combined with proper embedding optimization and index configuration. You scale semantic search by implementing robust monitoring, performance tuning, and structured reindexing plans to maintain retrieval speed.

When should I use pre-filtering versus post-filtering in vector databases?▼

Use pre-filtering to restrict the search space before vector retrieval, improving speed with high-cardinality metadata. Use post-filtering to apply strict metadata constraints after retrieval, ensuring precise semantic search matches without excluding relevant vectors from the index.