Scale examples: DoorDash lift

Dr. Jure Leskovec is the Chief Scientist at Kumo.AI and a Stanford professor, working on relational foundation models and graph-transformer systems that bring enterprise databases into the foundation-model era.

Relational Foundation Models: Unlocking the Next Frontier of Enterprise AI // MLOps Podcast #348 with Jure Leskovec, Professor and Chief Scientist, Stanford University and Kumo.AI.

Join the Community:

https://go.mlops.community/YTJoinIn

Get the newsletter: https://go.mlops.community/YTNewsletter

// Abstract

Today’s foundation models excel at text and images—but they miss the relationships that define how the world works. In every enterprise, value emerges from connections: customers to products, suppliers to shipments, molecules to targets. This talk introduces Relational Foundation Models (RFMs)—a new class of models that reason over interactions, not just data points. Drawing on advances in graph neural networks and large-scale ML systems, I’ll show how RFMs capture structure, enable richer reasoning, and deliver measurable business impact. Audience will learn where relational modeling drives the biggest wins, how to build the data backbone for it, and how to operationalize these models responsibly and at scale.

// Bio

Jure Leskovec is the co-founder of Kumo.AI, an enterprise AI company pioneering AI foundation models that can reason over structured business data. He is also a Professor of Computer Science at Stanford University and a leading researcher in artificial intelligence, best known for pioneering Graph Neural Networks and creating PyG, the most widely used graph learning toolkit. Previously, Jure served as Chief Scientist at Pinterest and as an investigator at the Chan Zuckerberg BioHub. His research has been widely adopted in industry and government, powering applications at companies such as Meta, Uber, YouTube, Amazon, and more. He has received top awards in AI and data science, including the ACM KDD Innovation Award.

// Related Links

Website: https://cs.stanford.edu/people/jure/

https://www.youtube.com/results?search_query=jure+leskovec

Please watch Jure's keynote:

https://www.youtube.com/watch?v=Rcfhh-V7x2U

~~~~~~~~ ✌️Connect With Us ✌️ ~~~~~~~

Catch all episodes, blogs, newsletters, and more: https://go.mlops.community/TYExplore

Join our Slack community [https://go.mlops.community/slack]

Follow us on X/Twitter [@mlopscommunity](https://x.com/mlopscommunity) or [LinkedIn](https://go.mlops.community/linkedin)]

Sign up for the next meetup: [https://go.mlops.community/register]

MLOps Swag/Merch: [https://shop.mlops.community/]

Connect with Demetrios on LinkedIn: /dpbrinkm

Connect with Jure on LinkedIn: /leskovec

Timestamps:

[00:00] Structured data value

[00:26] Breakdown of ML Claims

[05:04] LLMs vs recommender systems

[10:09] Building a relational model

[15:47] Feature engineering impact

[20:42] Knowledge graph inference

[26:45] Advertising models scale

[32:57] Feature stores evolution

[38:00] Training model compute needs

[42:34] Predictive AI for agents

[45:32] Leveraging faster predictive models

[48:00] Wrap up