For Philip Dee, the pivot to a career in science was not a singular "aha" moment; it was a slow-building momentum. "I have always been generally interested in science ever since I was a child," Dee said. However, it was not until his high school introduced an experimental engineering program that he began to shape that interest into something more tangible. That early hands-on experience evolved into a deep fascination with physics during college, especially after a course in quantum mechanics and special relativity. "It was the first time I felt that sort of friction in my brain - something akin to a paradigm shift in perspective," he added. "I was deeply enthralled."
Dee is now a Eugene P. Wigner Distinguished Staff Fellow at Oak Ridge National Laboratory in the Computing and Computational Sciences Directorate. He develops machine learning-augmented simulations to explore quantum many-body systems. His work sits at the intersection of computational physics, quantum theory and emerging AI models, pushing the boundaries of how scientists study and predict the behavior of complex materials.
Dee was drawn to condensed matter physics because of its closeness to the types of experiments being conducted by scientists at the laboratory. "Condensed matter is interesting in that the iteration cycle is faster and more modular than, say, high-energy physics," he said. "Even if you are a theorist like I am, you can work adjacent to the people running the experiments, and you can talk to them regularly." That closeness became foundational to his research philosophy, which was cultivated during his postdoc years by the constant collaboration between theorists and experimentalists.
That collaborative mindset guides Dee's current research. One of his projects involves developing faster, more robust simulations of quantum systems by integrating ML into quantum Monte Carlo methods which are problem-solving techniques that use repeated random sampling to obtain a numerical result. "What excites me about it is that it is so open and active. There is a strong possibility that no one has tried the idea that pops into your head," he said. Although ML is not a cure-all, Dee is focused on making it more general, more robust and capable of accelerating simulations of quantum materials and interpreting results without sacrificing scientific accuracy.
A second branch of his work connects even more directly to real-world applications. By building a codebase of theoretical approaches that interfaces with ab initio methods, Dee aims to support experimental efforts, including those at ORNL's Spallation Neutron Source and the Center for Nanophase Materials Sciences, both Office of Science user facilities. He said, "We hope that by developing this suite of simulation tools, we will help experimentalists interpret their neutron scattering spectra, thereby helping to disentangle whether what they're seeing in the material is due to phonons, magnetic excitations or other microscopic details."
The journey has not always been smooth. One early project he initiated during his doctoral studies was scooped up by another group just as he was preparing to publish. "It is not a good feeling, but we ended up putting a spin on the problem and taking a more sophisticated route," he said. "I probably learned a lot more in the end."
That resilience has become part of his outlook on research: "You must have a sense of when the effort is meaningful. Sometimes progress slows, and ideas don't work out, but that is just a part of doing science." He credits his mentors - particularly those who straddled theory and experiment - with shaping the other aspects of his approach. "They had seemingly mastered the balance of being plugged into both worlds, staying aware of recent progress on important open questions, and that inspired me," he added.
Outside the lab, Dee's creative side continues to influence his scientific mind. A longtime musician, he sees parallels between improvisational music and theoretical physics: "In both cases, trying things and exploring are fundamental to making something new. Improvisational music teaches you to collaborate, adapt and take creative risks, and so does science.
For early career researchers, Dee's advice is both practical and philosophical: "Find the intersection between fields. That is where many open questions live. And once you find something interesting there, dive deep."
At the intersection of ML and quantum theory, Dee is not just solving equations; he is building bridges across disciplines, uncovering the hidden structures of matter and inspiring new ways to imagine what scientific discovery can look like.
UT-Battelle manages ORNL for DOE's Office of Science, the single largest supporter of basic research in the physical sciences in the United States. DOE's Office of Science is working to address some of the most pressing challenges of our time. For more information, visit energy.gov/science. - Neil Gillette
