NODES is the biggest graph community gathering dedicated to applications, data intelligence, knowledge graphs, and AI.
Let's find out what happened this year.

NODES, the online event on everything Graph and hosted by Neo4J, this year took place on November 6. This was the 7th occurrence of the conference and with a 24-hour live streaming session it hosted 160 talks from 200 speakers from all over the world.

It had everything for everyone. All skill levels were welcome.
The live sessions were divide by timezone; Asia Pacific, Europe and the Americas. Of course for someone to watch that amount of talks on that short of time window would be something unfeasible. But now thankfully all the talks have been uploaded to Youtube for everyone to enjoy. We've picked a few of our liking to highlight.

The top of our list is occupied by "Fireside Chat with Andrew Ng and Emil Eifrém". Two well known figures of the ML world, Andrew Ng and Emil Eifrem, CEO and Co-Founder of Neo4j, talk about the current and future status in the land of Artificial intelligence. If you want to get a glimpse of what the future holds then this is a must watch.

A very interesting talk was "CyRAG: Retrieval Augmented Generation for Robust Cyber Defense" which blends cyber security vulnerability discovery and alerting with LLM and RAG, basing on Neo4j knowledge graphs. This is demonstrated with CyRag, a defense tool work in progress that reduces the time between vulnerabilities discovered and fixing them.

Next on our list is "RAG and Voice Agent Applications" where Tony Xavier demonstrates how one can make intelligent voice agents.Making Voice agents is much more difficult than other kind of agents because they require what is described as "perfect memory" that understands context and can evolve over time.

Without this capability, the agent may ask the same questions over and over again, which is degrading in performance. So Tony suggests a solution that is a blend of Hybrid RAG and Graph Databases for context, plus specific data handling techniques in an attempt to overcome this memory issue and instead offer persistence.

Finally, a boon for all developers out there is "Leveraging Knowledge Graphs and LLMs to Document Large Scale Codebases". It's an issue in ever developer's mind.

• Asked to modify a legacy code base? Check.
  
  
• Ask senior devs to onboard junior devs in the current codebase? Check.
  
  
• Trying to understand a tutorial? Check again!

Lockheed Martin mostly famous for its fighter jets, is also a software house (aren't they all nowadays?) which has got the solution to the problem. They leverage Knowledge Graphs and Generative AI to automate software architecture documentation to create a structured representation of the code. That way they reduce errors and improve maintainability, rendering tasks like on-boarding a simple case.

To sum it up, this conference offered a grand opportunity for everyone to find out what Graph based models are and how their combination with LLMs and RAG can realize many of nowadays ML tasks.

More Information

NODES 2025 Youtube Playlist

Related Articles

An Introduction to Neo4j

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info