In this episode of the Kronos Fusion Energy Podcast, we are joined by Jack Dongarra, one of the foundational architects of modern high-performance scientific computing and the 2021 winner of the ACM A.M. Turing Award, the highest honor in computer science.
Jack Dongarra is internationally recognized for his pioneering contributions to numerical linear algebra and high-performance computing, including the development of software libraries such as LINPACK and LAPACK, which form the mathematical backbone of much of today’s scientific computing ecosystem and tools like MATLAB. His career spans decades at the intersection of mathematics, computer science, and large-scale scientific simulation, with leadership roles across U.S. national laboratories and the Department of Energy computing programs.
In this conversation with Priyanca Ford, Founder & CEO of Kronos Fusion Energy, Jack reflects on his personal journey—from early academic challenges to shaping the foundations of modern scientific computing—and explains how computation has evolved from supporting infrastructure into a true scientific instrument. The discussion explores why most supercomputers fail to reach their theoretical performance, the importance of co-design across algorithms, hardware, and applications, and how AI complements—but does not replace—physics-based numerical methods.
As a Scientific Advisor to Kronos Fusion Energy, Jack also discusses what it takes to build trustworthy, predictive simulations for safety-critical systems, including verification, validation, uncertainty quantification, and the role of digital twins in fusion energy development.
This episode covers:
- High-performance computing and numerical algorithms
- AI and physics-based simulation
- Supercomputer architecture and energy costs
- Co-design and future computing systems
- Fusion energy digital twins and scientific validation
This is both a masterclass in scientific computing and a deeply human conversation about curiosity, resilience, and the people behind the code.