GraphGeeks Podcast

Graph Chat with Golven Leroy on Predicting Graph Costs & Moving Past "It Depends"

How do we move past the ultimate graph answer: "It depends"? Live from the Open Data Science Conference (ODSC), Amy Hodler sat down with Golven Leroy, graph researcher and data science lead at Graphable, to discuss groundbreaking core mathematics that could change how we calculate the compute cost, latency, and bounds of graphs and trees—independently of technology. We also dive into his top takeaways from ODSC, including tracking LLM experiments with MLflow, revolutionary chip/circuit design, and open-source frameworks like Agor for visualizing enterprise AI agents. Link to the referenced paper: https://msp.org/involve/2026/19-2/p05.xhtml

Geometry vs. Topology in AI: Roie Schwaber-Cohen (Pinecone) & Amy Hodler

Is the future of AI fluid and semantic, or rigid and structured? In this episode of Graph Geeks in Discussion, host Amy Hodler sits down with Roie Schwaber-Cohen, Head of DevRel at Pinecone. Together, they explore a fascinating philosophical and operational debate: the intersection of vector geometry (semantic distance) and graph topology (explicit connectivity). Tune in to discover why solving the next generation of AI challenges, from hallucinations to agentic long-term memory, requires fusing both worlds, and why we all might just need to "embrace the fuzziness."

Memgraph Zero & Federated GQL with Marko Budiselić

Is data copying dead? In this episode of the GraphGeeks Podcast, host Amy Hodler sits down with Marko Budiselić ("Buda"), Co-founder & CTO at Memgraph, to discuss their major new releases: Memgraph Zero and MemGQL. Discover how Memgraph is tackling the massive pain point of ETL pipelines by creating a federated GQL layer that queries data directly at the source—across Postgres, MySQL, Neo4j, ClickHouse, and more. Buda also shares how a centralized semantic layer is becoming essential for the future of AI agents. Key topics covered: * The core architecture and vision behind Memgraph Zero. * How MemGQL acts as a federated graph query engine. * Moving beyond context graphs into environmental graphs for AI agents. * The shift toward agentic data modeling and mapping. 👉 Read more about the release: https://memgraph.com/docs/memgraph-zero 👉 Vote on the next data connectors in the Memgraph Community Roadmap Poll: forms.gle/2MLfWp24uwbJpsey8

Graph Chat with William Lyon on AI Agents, Graph Memory, and the Return to Neo4j

In this chat, Amy Hodler catches up with William Lyon to discuss his recent return to the Neo4j Product Team and his deep dive into the world of AI agent frameworks and memory. Wil discusses his hands-on workshop exploring a fascinating frontier: Graphs as the Memory for AI Agents. Moving beyond simple retrieval, Will explains how we can use graph technology to mirror human cognitive functions, including: * Episodic Memory: Learning from user interactions. * Semantic Memory: Building a canonical model of the world. * Procedural Memory: Unlocking advanced, graph-based reasoning for agents. Whether you're interested in the latest in GraphRAG, the evolution of Knowledge Graphs in the age of LLMs, or just want to hear about Will’s journey through the startup ecosystem, this conversation is packed with insights that have only become more vital since they were recorded.

Graph Chat with Paco Nathan on AI as a Practice, Not a Product

Bryce Merkl Sasaki (Head of Marketing at gdotv) sits down with the "Gandalf of Graph Technology" himself, Paco Nathan (Senior DevRel at Senzing), at the Open Data Science Conference. Paco is a pioneer in neural networks and NLP since the 1980s, and a leading voice in the MLOps and Graph communities. In this chat, Paco cuts through the current hype bubble to discuss how graph technologies and entity resolution are solving high-stakes, real-world problems. From interdicting $3 trillion in dark money and human trafficking to fixing systemic data disconnects in government agencies through the NIEM semantic standard, this conversation explores the profound human impact of well-engineered data. Key Topics: * Entity Resolution (ER): How connecting data points protects real people in the legal and financial systems. * Semantic Standards: Using NIEM and SKOS for better context engineering * Why the biggest bottleneck in AI isn't the math—it's designing UX & visualizations that humans can actually use. * The AI Bubble: Paco’s perspective on the current industry jitters vs. the tangible tech that is ready for production.