video1.0<iframe src="https://www.loom.com/embed/f9f7521ec9c9478088761c00195ea781" frameborder="0" width="1862" height="1396" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>13961862Loomhttps://www.loom.com13961862https://cdn.loom.com/sessions/thumbnails/f9f7521ec9c9478088761c00195ea781-eb1ac6cd95a1fe32.gif108.563This Loom demonstrates how to turn AI notes into a knowledge graph and use it to power an LLM-based chatbot. The speaker uses an LLM with open-source libraries like PyVis to create an interactive graph showing concepts and relationships, such as Attention Mechanism powering Transformer Architecture, which in turn powers large language models. The chatbot understands a user question, generates a SPARQL query to retrieve relevant graph concepts and relationships, and then summarizes them into a response. The interface also highlights the extracted sub-graph and provides a dropdown for graph context showing detailed relationships and concept descriptions. The demo is built on a couple of notes but is intended to scale to hundreds of thousands of demos in production.