Graph Chat with Chang She, CEO & Co-Founder of LanceDB

Times are changing: Back in my day, querying and organizing your graph data required a graph database. rocking chair squeaks

But these days, the young folks are more often opting for a graph query engine to analyze the hidden connections and relationships in their data without storing it as graph data. They say that sometimes it’s all they need. (Of course, many use cases totally still do require a graph database, I reckon!)

One of those new graph query engines didn’t start off as a graph project at all, but now its expanding frontiers have crossed over with the world of graph technology in new and interesting ways. That project is LanceDB.

In this Graph Chat video series for GraphGeeks, community member David Hughes interviewed Chang She, CEO and Co-founder of LanceDB, at ODSC West in San Francisco. Catch the video below!

Graph Chat Interview with Chang She, CEO & Co-Founder of LanceDB

In this Graph Chat, Chang She shares groundbreaking insights on how agentic retrieval systems are challenging traditional RAG approaches, requiring much higher throughput and iterative search. The conversation highlights the new Lance Format as the Multimodal Lakehouse standard optimized for AI data operations.

Most exciting for the graph community, Chang introduces the new open source project, Lance Graph, which enables storing graph schemas and executing Cypher-like queries directly on Lance tables, integrating vector, tabular, and graph data into a unified format.

David and Chang wrap-up their convo talking about why data differentiation is the key to winning in the age of AI agents.

More about LanceDB & Lance Graph

LanceDB is the multimodal AI lakehouse built on top of the Lance columnar format. It’s designed for fast, scalable, and production-ready vector search. Think of LanceDB as a central location where you can build, train and analyze your AI workloads – storing, indexing, and searching over petabytes of multimodal data and vectors with ease.

Lance Graph is a Cypher-capable graph query engine built in Rust with Python bindings for building high-performance, scalable, and serverless multimodal knowledge graphs.

Subscribe for More Graph Chats

The Graph Chat video series by GraphGeeks aims to ask diverse leaders and practitioners about where the future of graph technology is headed. This and other episodes were sponsored by gdotv – the graph database IDE. Check out the full playlist of Graph Chats for more!

Special thanks to David Hughes (On-Screen Talent, Tech Wizard), Amy Hodler (Studio Exec, Producer), and Bryce Merkl Sasaki (Cameraman).

If you enjoyed this Graph Chat interview with Chang She, subscribe to the GraphGeeks YouTube channel for more great discussions with leaders across the graph technology space.