Graph Pulse: Dr. Clair Sullivan, Graph Data Scientist & Consultant

追加: 2026-06-03

[00:00:00]Hi, my name is Claire Sullivan. I am presently running my own company doing consulting, coding, and education in the GenAI space.

[00:00:13]My work at Neo and pre prior to that was in both graphs and natural language processing. And so this was pre, you know, things like ChatGPT going GA. And when that happened, you know, these are all the mathematical basis for what is LLM today. So, I work with a lot of people who are looking, you know, for traditional graph work, but also they're looking for LLM work and you know, how those two things come together. So, um for example, I've recently taught a course on LinkedIn Learning about GraphRAG um with, you know, so graphs LLMs, answering questions, making for better RAG over vector-based approaches.

[00:00:56]Um I do a lot of work with my clients on that. Um and I and I even do still some very traditional graph work. Just really, you know, understanding data and traversals within data and what is much easier to do within a graph and much more possible than something like, you know, a gazillion SQL tables and joins and you know, how slow those can be. So, um so it's really still spending the space of graphs all the way through GenAI and kind of the new stuff. As, you know, 5 years ago we were talking a lot about graph data science and things like that. And now, you know, graphs are really having a moment, but that moment that they're having really isn't the traditional space if you think about it because when we talk about LLMs and GenAI and GraphRAG, those are really just traversal things. They're not complicated. But instead of traversing, you know, things like customers or products or things like that, you know, now we're traversing a knowledge graph.

[00:01:53]So, whereas 5 years ago I think people just started maybe hearing about knowledge graphs and now today like if you're involved at all in gen AI, you've heard of knowledge graphs, you've heard of graph rag and it's really I think a very exciting time in graphs because we're getting a lot of new people coming into the space, people with great new ideas who are you know, have these wonderful applications that they're looking at creating within gen AI all using graphs. I think people who are successful in the industry will understand those fundamentals, how to really apply proper traversal techniques and storage techniques and and the ones who are able to figure out how to bring more people on board in an easy way. Graphs for a long time have been seen as niche, but now they're not. Now they are something that everybody really needs to be involved in, take seriously, learn how to use. So, what I see, you know, this is I can't predict the future, but if I could, I would imagine that it would look something like how do we make it make graphs easier for more people to use.