• “tech”
  • “ai”
  • “rag”
  • “knowledge-graph”
  • “neo4j”
  • “ai-tutorials”
  • “rag-systems”
  • “knowledge-graphs”
  • “graph-rag”
  • “light-rag” aliases:
  • “Tech With Homayoun Tutorials”

Tech with Homayoun

Practical AI and tech tutorials focusing on implementable systems, particularly rag and knowledge-graph applications.

Key Concepts

• Light RAG vs Graph RAG: Hybrid approach using neo4j for knowledge graph storage alongside vector stores, contrasting with pure Graph RAG implementations
• Core Architecture:
  • Document chunking
  • Node/relationship extraction
  • Dual storage in vector store and neo4j knowledge graph
  • LLM response generation using structured context
• Benefits: Efficiency in retrieval while maintaining relationship context (vs. pure vector search)

Recent Notes

• 2026 04 14 Build a light RAG system with neo4j
  • Demonstrates implementation using neo4j for knowledge graph storage
  • Contrasts Light RAG with Graph RAG
  • Covers document chunking → node extraction → dual storage in vector store and neo4j
  • Explains overall architecture of LightRAG framework
  • Includes video tutorial: https://www.youtube.com/watch?v=zR9I7aMI8vw