# Day 1 Lecture: Graph RAG vs Vector RAG

> In this lecture, Ashwani provides an overview of vector retrieval augmented generation (RAG) and introduce knowledge graphs, highlighting their evolution and significance in enhancing search capabilities. He discuss the limitations of vector RAG, particularly in handling deep relationship queries a…

- Web page: https://www.dataexpert.io/lesson/advanced-rag-and-agentic-ai-day1-lecture-july2025-p737-l1178
- Program: [AIExpert](https://www.dataexpert.io/program/ai-expert)
- Module: Week 2: Advanced RAG, Reranking & Dev Workflow
- Access: Requires enrollment in AIExpert
- Length: 39 min video
- Skills: NLP, LLMs, Data Modeling, Data Structures, Algorithms
- Academy: DataExpert.io Academy

## About this lesson

In this lecture, Ashwani provides an overview of vector retrieval augmented generation (RAG) and introduce knowledge graphs, highlighting their evolution and significance in enhancing search capabilities. He discuss the limitations of vector RAG, particularly in handling deep relationship queries and fragmented context, and how knowledge graphs can address these challenges. He also explains the process of constructing knowledge graphs from structured and unstructured data, emphasizing the importance of defining schemas for effective graph creation. Lastly, he outlines the retrieval patterns and strategies for leveraging knowledge graphs.

## Navigation

- Previous lesson: [Optimize your dev workflow with Claude Code and Codex Day2 Lab](https://www.dataexpert.io/lesson/optimizeyourdevworkflowwithclaudecodeandcodexday2lab-mar26-p737-l2144.md)
- Next lesson: [Day 1 Lab: Building Graph RAG in production](https://www.dataexpert.io/lesson/advanced-rag-and-agentic-ai-day1-lab-july2025-p737-l1179.md)