S3 EP9 - Fluid Intelligence with Johannes Brandstetter and Siddhartha Mishra

In this conversation, Neil Ashton and Prof. Siddhartha Mishra, and Prof. Johannes Brandstetter discuss their recent paper on AI foundation models in computational fluid dynamics (CFD). They explore the backgrounds of the speakers, the journey to writing the paper, the role of AI in CFD, and the challenges of scaling laws and data generation. The discussion also covers model training costs, open questions, and future directions for research in this field.

Fluid Intelligence: A Forward Look on AI Foundation Models in Computational Fluid Dynamics : https://arxiv.org/abs/2511.20455v1

Spotify version: https://open.spotify.com/episode/48Om...

Chapters

00:00 Introduction and Setup
02:54 Background of the Participants
05:56 The Journey to the Paper
08:38 Understanding the CFD Process
11:51 The Role of AI in CFD
14:37 Exploring the Paper's Structure
17:50 Input-Output Relations in CFD
21:01 Quantifying Computational Costs
23:47 CFD vs. Language Models
26:40 Use Cases of AI in CFD
29:37 Challenges in Turbulence Modeling
32:43 Scaling Laws and Hypotheses
35:41 Final Thoughts and Future Directions
40:19 Data Generation and Cost Analysis
41:38 Modeling Paradigms and Their Impact
42:34 Transient Data Utilization
45:17 Sampling Strategies in Transient Simulations
48:32 Downsampling and Its Implications
51:33 Model Training and Computational Costs
55:22 Scaling Laws and Data Generation
01:00:39 Open Questions in Model Training
01:08:58 Inductive Bias and Physics Integration
01:19:12 Future Directions and Industry Applications