RAG Needs a Map: Using GraphRAG to Retrieve Connected Context — Nyah Macklin, Neo4j
Source
youtube.com
Author
AI Engineer
Date
Why it matters
Vector search returns similar chunks but not relationships. A schema-guided knowledge graph adds traversal and aggregation, and the demo shows why to verify the generated query behind an answer.
Key takeaways · AI-distilled
Macklin's pipeline has an LLMA large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.Full definition → extract entities and relationships from course material, embeddingA list of numbers representing a piece of text's meaning, so that similar meanings end up numerically close and can be searched.Full definition → handle similarity search, and Cypher exposes the resulting structure, so the graph adds traversal and aggregation on top of vectors rather than replacing them.
The AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → built in Neo4j Aura combines a vector index, a lesson search tool whose retrieval query also returns related graph data, and text-to-Cypher for natural language questions, and it shows the sources behind each answer.
The code walkthrough defines chunkingSplitting documents into passages small enough to embed and retrieve individually — the step that quietly determines whether retrieval works at all.Full definition →, entity types and relationship types up front, so a schema guides extraction instead of leaving the model to invent the graph's structure.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
embedding — A list of numbers representing a piece of text's meaning, so that similar meanings end up numerically close and can be searched.
chunking — Splitting documents into passages small enough to embed and retrieve individually — the step that quietly determines whether retrieval works at all.
videoAI Security Engineer Foundations + Certificate — Micah Silverman, Snyk
videoThe SDLC You Build Yourself — Andy Wong, Andrei Bocan & Shane Wolf, Atlassian
videoAgent Speedrun — Elizabeth Fuentes Leone & Sandhya Subramani, AWS
videoCan Your Agent Hear You Now? Building Live Voice Agents with Gemini — Thor Schaeff
