# Knowledge Graph Engineering and GraphRAG. Canonical: https://brew.new/browse/templates/email/pt1_k97yet6d70vpgxenjqv29pe9ys8e37p2 Brand: substack.com Category: newsletter ![Preview of Knowledge Graph Engineering and GraphRAG.](https://cdn.brew.new/email-preview-bd2fd835ab91d716-tracking_r57xw9fw8weyne6kjtrgfe6h9x8e2fsz-1788974090454.png) ## Email content Forwarded this email? Subscribe here for more Knowledge Graph Engineering and GraphRAG. Your AI Finds the Files. Can It Connect the Dots? ChatGPT Sep 9 ∙ Preview READ IN APP After our articles on AI Foundations, Prompt Engineering, Context Engineering, Memory Engineering and RAG, we have reached the sixth part of our practical AI series: Knowledge Graph Engineering and GraphRAG. My promise with this series is to help you build skills you can actually use. Each article should leave you with something working, something you understand and something you can improve. Today, we will extend the assistant from our RAG article. You probably know the frustration of finding the right document and still having to work out the answer yourself. A deadline changes. You find the update, but now you need to remember which tasks depend on it, who owns them and whether those assignments are still current. The information exists. Connecting it is still your job. That is the problem we will work on here. Our previous assistant searched for relevant passages. This time, we will give it explicit relationships to follow, with a source behind each connection. That structure is a knowledge graph. You will see this through the workshop project we have been building together. We will trace how a supplier delay affects other work, identify the people responsible and change an owner to check whether the assistant follows the update. If you missed the RAG article, you can read it here. Today’s download also includes the earlier records, so you can follow along without rebuilding everything. The question is simple: can your assistant connect the relevant information and show you enough evidence to trust its answer? In this article, you’ll build a small dependency graph, retrieve its supporting evidence and test an approved ownership change.You’ll get the runnable project, comparison results and step-by-step checks to publish your work on GitHub... User's avatar Continue reading this post for free in the Substack app Claim my free post Or upgrade your subscription. Upgrade to paid Like Comment Restack © 2026 ChatGPT 548 Market Street PMB 72296, San Francisco, CA 94104 Unsubscribe Get the appStart writing [Open and remix this design](https://brew.new/browse/templates/email/pt1_k97yet6d70vpgxenjqv29pe9ys8e37p2) [Browse email designs](https://brew.new/browse/templates)