S3 EP4 - 5 tips for CAE engineers in the era of AI
In this episode of the Neil Ashton podcast, Neil discusses the impact of AI on CAE engineering, providing five essential tips for engineers to thrive in this evolving landscape. The conversation covers the importance of maintaining an open mind, continuous education, and preparing for AI physics applications. It also delves into the build vs. buy dilemma for AI solutions and the emerging concept of agentic AI, which promises to revolutionize engineering practices.Chapters00:00 Introduction to the Podcast and AI in Engineering01:03 Five Tips for CAE Engineers in the Era of A101:24 1: Keeping an Open Mind 07:39 2: Understanding AI Physics and Its Applications13:30 3: Preparing for AI Implementation in Engineering18:54 4: The Build vs. Buy Dilemma in AI Solutions22:20 5: The Future of Agentic AI in Engineering
-------- Â
23:48
--------
23:48
S3 EP3 - Professor Johannes Brandstetter on AI for Computational Fluid Dynamics
In this conversation, Neil Ashton interviews Prof. Johannes Brandstetter, a physicist turned machine learning expert, about his journey from academia to industry, focusing on the application of machine learning in engineering and computational fluid dynamics (CFD). They discuss the Aurora project, the challenges of integrating machine learning with engineering, and the importance of data in training models. Johannes shares insights on the use of transformers in modeling, the significance of resolution independence, and the role of open-source practices in advancing the field. The conversation also touches on the challenges of founding a startup and the need for multidisciplinary collaboration in tackling complex engineering problems.Links:Â Github: https://brandstetter-johannes.github.ioEmmi AI: https://www.emmi.aiGoogle scholar: https://scholar.google.com/citations?user=KiRvOHcAAAAJ&hl=deAB-UPT transform paper: https://arxiv.org/abs/2502.09692Chapters00:00 Introduction to Johannes Brandstetter07:10 The Aurora Project and Key Learnings11:15 Machine Learning in Engineering and CFD17:19 Challenges with Mesh Graph Networks20:16 Transformers in Physics Modeling31:14 Tokenization in CFD with Transformers39:58 Challenges in High-Dimensional Meshes41:08 Inference Time and Mesh Generation41:36 Neural Operators and CAD Geometry45:59 Anchor Tokens and Scaling in CFD48:40 Data Dependency and Multi-Fidelity Models50:32 The Role of Physics in Machine Learning54:28 Temporal Modeling in Engineering Simulations56:58 Learning from Temporal Dynamics1:00:58 Stability in Rollout Predictions1:03:48 Multidisciplinary Approaches in Engineering1:05:18 The Startup Journey and Lessons Learned
-------- Â
1:18:02
--------
1:18:02
S3 EP2 - Prof. Russell Cummings - World leader in Aerospace Engineering and Hypersonics
In this episode of the Neil Ashton podcast, Professor Russell Cummings shares his extensive journey through the fields of aerodynamics, computational fluid dynamics and hypersonics. He discusses his early inspirations, his early days at University and the Hughes Aircraft Company - a key time during this life. He also talks about the cyclical nature of hypersonics research, and the challenges faced in computational fluid dynamics (CFD). Prof. Cummings emphasizes the importance of perseverance in engineering careers and the need for collaboration between experimental and computational methods. He also shares insights on the role of AI in hypersonics and offers valuable advice for aspiring engineers.Prof. Russ Cummings graduated from California Polytechnic State University (Cal Poly) with a B.S. and M.S. in Aeronautical Engineering, before receiving his Ph.D. in Aerospace Engineering from the University of Southern California; he also received a B.A. in music from Cal Poly. He is currently Professor of Aeronautics at the U.S. Air Force Academy and Director of the Hypersonic Vehicle Simulation Institute. Prior to this he was Professor of Aerospace Engineering at Cal Poly, where he also served as department chairman for four years. He also worked at Hughes Aircraft Company, and completed a National Research Council postdoctoral research fellowship at NASA Ames Research Center, working on the computation of high angle-of-attack flowfields. He is a Fellow of the Royal Aeronautical Society and the American Institute of Aeronautics and Astronautics.Distribution Statement A: approved for public release, PA# USAFA-DF-2025-652. The views expressed in this interview are those of the author and do not necessarily reflect the official policy or position of the United States Air Force Academy, the Air Force, the Department of Defense, or the U.S. Government.LinksAerodynamics for engineers: https://www.cambridge.org/us/universitypress/subjects/engineering/aerospace-engineering/aerodynamics-engineers-7th-edition?format=HB&isbn=9781009501309RAeS Lanchester Named Lecture 2024: Frederick W. Lanchester and 'Aerodynamics' https://www.youtube.com/watch?app=desktop&v=lApNzYaZOmk&t=884s NASA at 50 (Prof Cummings is in the picture): https://images.nasa.gov/details/ARC-1989-AC89-0276-6 Chapters00:00 Introduction to the Podcast and Guest04:56 Professor Russell Cummings: A Journey Through Engineering31:14 The Evolution of Hypersonics Research58:26 The Role of AI in Hypersonics and CFD01:37:55 Advice for Aspiring Engineers
-------- Â
1:43:14
--------
1:43:14
S3 EP1 - Prof. Mike Giles - A CFD and Computational Finance Pioneer
In this episode of the Neil Ashton podcast, Professor Mike Giles shares his extensive journey through the fields of computational fluid dynamics (CFD), computational finance and HPC. He discusses his early academic influences, his early days at Cambridge, internships at Rolls-Royce, his transition to MIT and Oxford where he made significant contributions to high-performance computing and numerical analysis. The conversation highlights his hands-on approach to research and teaching, as well as his pioneering work in Monte Carlo methods and GPU computing. This conversation explores the journey of a mathematician and engineer from MIT to Rolls-Royce and then to Oxford, highlighting the evolution of computational engineering, the development of the Hydra code, and the transition from CFD to financial applications. In this conversation, the speaker reflects on their journey through burnout, career transitions, and the evolution of their work in computational finance and numerical analysis. They discuss the challenges of managing large software projects, the shift from Hydra code development to finance, and the integration of advanced methodologies in their work. The conversation also touches on the role of high-performance computing, the impact of AI on research, and advice for the next generation of students pursuing careers in mathematics and programming.Links:https://people.maths.ox.ac.uk/gilesm/Chapters00:00 Introduction 06:25 Professor Mike Giles: A Journey Through CFD and Finance17:30 Early Academic Influences and Career Path29:34 Transition to MIT and Early Research40:01 High-Performance Computing and Its Impact41:30 Navigating Between MIT and Rolls-Royce44:54 The Evolution of Research at MIT48:47 Transitioning to Oxford and the Role of Rolls-Royce51:07 The Genesis of the Hydra Code01:00:47 The Role of Conferences in Engineering01:10:58 The Shift from CFD to Financial Applications01:21:30 Navigating Burnout and Career Transitions01:24:04 Shifting Focus: From Hydrocode to Computational Finance01:29:30 Bridging Mathematics and Finance: Methodologies and Techniques01:35:09 The Role of High-Performance Computing in Modern Research01:39:20 AI's Impact on Research and Future Directions01:54:00 Advice for the Next Generation: Pursuing Passion and Skills
-------- Â
2:07:11
--------
2:07:11
S2 EP11 - Foundational AI Models for Fluids
In this episode of the Neil Ashton podcast, the discussion revolves around foundational models in fluid dynamics, particularly in the context of computational fluid dynamics (CFD). Neil shares insights from a recent panel discussion and explores the potential of AI in predicting fluid behavior. He discusses the evolution of AI in CFD, the challenges of data availability, and the differing adoption rates between industries. The episode concludes with predictions about the future of foundational models and their impact on the engineering landscape.Chapters00:00 Introduction to the Podcast and Topic01:09 Foundational Models in Fluid Dynamics10:09 The Evolution of AI in CFD19:52 Future Predictions and Industry Dynamics
This podcast focuses on explaining the fascinating ways that science and engineering change the world around us. In each episode, we talk to leading engineers from elite-level sports like cycling and Formula 1 to some of world's top academics to understand how fluid dynamics, machine learning & supercomputing are bringing in a new era of discovery. We also hear life stories, career advice and lessons they've learnt along the way that will help you to pursue a career in science and engineering.
South Africa