04 Dec 2025
47m

The AI Revolution Finally Comes to Structured Data

Podcast cover

The Data Exchange with Ben Lorica

Kumo.ai's relational foundation model for enterprise data is the focus, with Jure Leskovec, Stanford professor and Kumo.ai co-founder, detailing its capabilities. Kumo.ai targets AI predictions and risk scoring over structured business data, offering an alternative to traditional data science teams or generic large language models. The relational foundation model uses a pre-trained, frozen model connected to a company's data via primary foreign key relations, enabling real-time predictive queries. This approach represents databases as graphs, using attention mechanisms over raw data to eliminate the need for manual feature engineering. Success stories include DoorDash's restaurant recommendations and Reddit's advertising models, both of which saw significant improvements using Kumo.ai.

Outlines

Part 1: Introduction, Current Challenges

Part 2: Relational Foundation Models, Core Technology

Part 3: Data Management, Model Refinement

Part 4: Advanced Features, Use Cases

Part 5: Ecosystem, Future Outlook

Sign in to continue reading, translating and more.

Open full episode in Podwise