This Week in Neo4j: Parallel Runtime, Geospatial, LangChain, RAG and more

Welcome to This Week in Neo4j, your weekly fix for news from the world of graph databases!
How cool was NODES2023? I know I am still thrilled about the many sessions and many interactions with lots of you throughout the day. I hope you had a great day as well and if you missed it: Don’t worry, we will share the recordings from the day with you soon.

Ahead of NODES, we also shared a few Neo4j product announcements: Parallel Runtime Capability (check out the hands-on blog below), Change Data Capture (CDC) and Easier Knowledge Graph Creation and more. Beyond that, we have an interesting geospatial app running on Neo4j, another LangChain article and we use RAG with Vector Search.

I hope you enjoy this issue,
Alexander Erdl


Stephen McGowan is a software developer with 15 years of experience. He has spent most of his career innovating in LegalTech and HealthTech, working with the oldest professions in the world and applying modern solutions.
Connect with him on LinkedIn.

Stephen was part of NODES 2023 where he explored the power of using domain-specific languages to create graphs. He showed how to use a domain-specific language for encoding regulation to create beautiful visualisations and rapidly deliver insights into complex topics.

Stephen McGowan

PARALLEL RUNTIME: Speed up your queries with Neo4j’s new parallel runtime

Parallel runtime is a fantastic alternative for using the available CPUs to execute graph-global read queries. In this article, Jens Pryce-Åklundh and Alex Averbuch briefly explain what “query runtime” means and then go a little deeper into some situations where the new parallel runtime shines.
GEOSPATIAL: U.S. Flu Vaccine Provider Search
Christopher Roberts built an app with Neo4j (lightning fast, because it relies on POINT INDEX), Streamlit and Folium for you to find the nearest point to get your flu shot in the United States by ZIP code.
LANGCHAIN: Constructing knowledge graphs from text using OpenAI functions: Leveraging knowledge graphs to power LangChain Applications
Tomaz Bratanic wrote another fantastic blog where he uses OpenAI functions with LangChain to build a knowledge graph from a Wikipedia page. You can follow his guide step by step in the article, but he also published the code on Github for you to repeat what he did.

RAG: Graphileon, Neo4j and ChatGPT RAGing together

This article delves into the intricacies of setting up a vector search index in Neo4j, creating nodes, and the potential of ChatGPT in generating context-aware responses. By the end, you’ll gain insights into the seamless integration of graph databases, vector search, and generative AI.


Don’t forget to share it if you like it!