The Neil Ashton Podcast

• S3 EP7 - 5 key trends for CFD revisited

  In this episode of the Neil Ashton podcast, the host revisits key trends in Computational Fluid Dynamics (CFD) from the past year, focusing on the rise of GPUs, advancements in AI and machine learning, the shift to cloud computing, the increasing adoption of high fidelity methods, and ongoing mergers and acquisitions in the industry. Each trend is explored in depth, highlighting the implications for the future of engineering and technology

  --------  

  20:08

  --------

  20:08
• S3 EP6 Prof. Brian Launder - CFD and Turbulence Modelling Pioneer

  In this episode, Professor Brian Launder (Professor at the University of Manchester and Fellow of the Royal Society and Royal Academy of Engineers) shares his remarkable journey through academia, detailing his early fascination with heat transfer, his transition to MIT, and his significant contributions to turbulence modeling and computational fluid dynamics (CFD). We touch upon the key role that Professor Brian Spalding had on his career as well as work that led to the breakthrough k-epilson turbulence model as well as the pioneering work on second-moment closure model. Prof Launder highlights the key role of collaborators and ex students such as Professors Hector Iacovides, Tim Craft, Bill Jones, Kemal Hanjalić and many more. He ends with advice for early-stage researchers and reflections on more than 50 years worth of academic research.Chapters00:30 Introduction05:00 Early Academic Journey10:06 Transition to MIT and Research Focus16:21 Return to Imperial College and Early Career21:06 Research Projects and PhD Students27:46 Development of the k-epilson model33:18 CHAM and Career Changes36:24 Move to UC Davis and New Research Directions44:05 Challenges and Opportunities in Research47:07 The Interview Experience51:14 Transition to Manchester University52:23 Research Innovations in Turbulence Modeling57:45 The Development of the TCL Model01:03:15 Nonlinear Eddy Viscosity Models01:05:58 Advanced Wall Functions and Their Applications01:10:09 Reflections on Career and Contributions01:15:49 Legacy and Impact on Turbulence ModelingTop Turbulence Modelling contributions (https://scholar.google.com/citations?user=Y3JbAK8AAAAJ&hl=en) 

  --------  

  1:28:23

  --------

  1:28:23
• S3 EP5 - Joris Poort - CEO and Founder of Rescale

  In this episode, Joris Poort, CEO and founder of Rescale, shares his personal journey on founding Rescale as well as his thoughts on the future of CAE. He discusses the challenges of introducing HPC to the cloud market, the traits that make successful founders, and the importance of perseverance and execution in entrepreneurship. Joris reflects on the early days of Rescale, the significance of early investors, and the evolving landscape of cloud computing and AI integration in engineering. The conversation highlights the complexities of transitioning to cloud solutions and the future potential of HPC in various industries. In this conversation, Joris discusses the transformative impact of AI on engineering, particularly in the context of inference, simulation, and automation. He emphasizes the importance of efficiency in engineering processes and how AI can significantly reduce the time required for complex simulations. The discussion also touches on the cultural shifts within organizations as they adapt to AI technologies, the potential for AI surrogates to revolutionize engineering practices, and the challenges of closing the sim-to-real gap. Joris offers insights for aspiring founders, encouraging them to pursue meaningful work that can drive innovation and societal progress.Chapters00:00 Introductions03:30 The Genesis of Rescale: A Cloud Computing Journey05:21 From Engineering to Entrepreneurship: The Leap of Faith09:28 Traits of a Successful Founder: Courage and Perseverance14:51 Tactical Steps to Startup Success: Building from the Ground Up22:10 Milestones and Breakthroughs: The Early Days of Rescale30:54 Navigating Challenges: The Role of Cloud Providers in HPC35:24 The Intersection of HPC and AI Training37:05 Cloud vs On-Premise: The Cost Debate39:54 Complexities of HPC in Enterprises42:27 The Slow Shift to Cloud Adoption44:34 Optimizing Workloads with Rescale46:50 Usability Challenges in Enterprise Software48:32 The Rise of Neo Clouds and Competition51:18 Speed and Efficiency in AI Training54:34 AI's Transformative Impact on Engineering58:54 The Future of AI Surrogates in Design01:03:28 Agentic AI: The New Paradigm in Engineering01:14:21 Solving Real Business Problems01:19:26 The Impact of AI on Engineering01:22:27 Innovation in Aerospace and Beyond01:25:19 Cultural Change in Organizations01:28:34 The Future of AI and Engineering01:39:09 Advice for Aspiring FoundersKeywordsHPC, cloud computing, startup journey, Rescale, entrepreneurship, AI, technology, innovation, engineering, business, AI, engineering, inference, simulation, automation, digital twin, innovation, aerospace, machine learning, technology

  --------  

  1:39:48

  --------

  1:39:48
• 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

  --------  

  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: 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

  --------  

  1:18:02

  --------

  1:18:02

More Science podcasts

Trending Science podcasts

About The Neil Ashton Podcast