Oreoluwa Alade shares story of his journey into computational physics and plans for the future in this interview with Racheal Olatayo

What inspired your journey into computational physics and how did machine learning become a core part of your work?

My journey into computational physics began during my undergrad at FUTA, where a course on computational physics sparked my interest. I’ve always loved math, and that course showed me how powerful simulations can be in solving real-world problems. Around 2020, the AI boom caught my attention, and I started teaching myself machine learning out of curiosity. It wasn’t just hype to me—it was a new way of thinking.

Over time, I started applying ML in real research, especially during my PhD in Physics at NDSU. For example, I developed algorithms to study microgel behavior and led a computer vision project focused on fisheye image detection. These weren’t just academic exercises—they had real implications in materials science and computer vision. I found that blending physics with ML allowed me to solve problems neither field could handle alone.

So what started with curiosity and a course has grown into a career where I’m constantly pushing boundaries, working on exciting problems, and applying AI to advance science in meaningful ways.

Can you explain your current research on compressible microgels and why it matters in soft matter physics and materials science?

My current research focuses on compressible microgels, which are tiny soft particles that can swell or shrink depending on their environment. What makes this work unique is that we’re the first to actually study the behavior of microgels as compressible—most past studies assumed they were fixed in size.
By introducing compressibility into the model, we’ve discovered that these microgels behave very differently than previously thought, especially in how they transition from fluid to solid states.

This matters a lot in soft matter physics and materials science because it gives a more realistic picture of how these soft particles interact, flow, and pack together under different conditions.

Beyond theory, this research could impact real-world applications like drug delivery systems, where the ability to shrink and expand can be used to release medication in a controlled way, or even in flexible optical materials that respond to pressure. So in summary, it’s a novel direction in the field, and I’m excited about the possibilities it’s opening up.

How have you applied machine learning techniques to problems in physics, and what insights have they revealed?
I’ve used machine learning to solve complex problems in soft matter physics. For instance, in my research on compressible microgels—a novel area where we study how soft particles change phase—I used machine learning models like random forest and linear regression to predict phase behavior with high accuracy. This revealed how factors like compressibility drive crystallization in materials, offering new insights for materials science.

I’ve also applied deep learning to correct distortion in fisheye images using convolutional neural networks and transformer-based models. These models significantly improved object detection accuracy in wide-angle images, which is useful in both scientific imaging and real-world applications like autonomous vehicles.

What’s a project you’ve worked on where your modeling or simulations significantly advanced understanding or outcomes?
One of the most significant projects I’ve worked on is my research into compressible microgels. These are soft, sponge-like particles that can swell or shrink depending on their environment. Traditionally, most models assumed that microgels were fixed in size, but I developed a novel computational approach that allows them to change size in response to crowding — something no one had done before.

Using Monte Carlo simulations, I studied how these microgels transition from a fluid to a solid state. I discovered that softness or compressibility plays a big role in their phase behavior. For example, softer microgels delay crystallization by shrinking in crowded spaces, which changes the stability and structure of the material. This has important applications in materials science and biomedical engineering, like in drug delivery systems or designing smart hydrogels.

The insights from this work help scientists better understand how soft materials behave under pressure and provide a roadmap for tuning the properties of these materials for real-world use.

What challenges have you faced while building models that bridge theory and real-world data?
One major challenge has been capturing the complexity of real-world behavior in models without oversimplifying or overfitting. In my microgel simulations, for instance, bridging theoretical physics with realistic particle behavior meant accounting for deformation, interpenetration, and crowding — all in one model. That’s not easy to do analytically, so I had to combine statistical physics with numerical simulations to get accurate, meaningful results.

Another challenge has been data availability. In computational physics, experimental data can be limited or noisy, so I’ve had to validate my models using physical principles or simulated benchmarks. Even with machine learning, ensuring that models generalize beyond training data requires careful feature engineering and regularization.

Ultimately, it’s about balancing realism with simplicity — making sure the model is physically sound but still usable and insightful.

You’ve presented at international conferences. Can you share what that experience was like and how it shaped your work?
Presenting at international conferences like the American Physical Society (APS) March Meeting and the International Soft Matter Conference (ISMC) has been a game-changer for me. These events exposed me to a global community of scientists working on cutting-edge research in physics and materials science.
Engaging with experts during my poster and oral presentations helped sharpen the way I communicate complex ideas, and I received valuable feedback that directly influenced how I approached key challenges in my microgel simulations.

More than anything, these conferences validated the significance of my work. They reminded me that what I’m doing isn’t just theoretical — it’s part of a larger conversation shaping the future of soft matter research.

What role has high-performance computing (HPC) played in scaling your research?
High-performance computing (HPC) has been essential to my research in both physics and AI. At North Dakota State University, I use the CCAST supercomputing cluster to run large-scale simulations and train complex deep learning models.
In my soft matter research, HPC lets me simulate the internal behavior of compressible microgels—something that requires tracking thousands of particles and their interactions over time. Without HPC, these simulations would take weeks or be impossible to run at the resolution I need. I also use molecular dynamics to study the atomistic structure of these materials, which would be computationally overwhelming without access to powerful resources.

On the AI side, I trained deep learning models for fisheye object detection—one training alone took 14 days using GPU nodes on the cluster. HPC made it possible to experiment, optimize, and improve model accuracy within a reasonable timeframe. It allowed me to build systems that work even in distorted image environments, which has wide applications in robotics and autonomous navigation.
HPC has helped me work at the intersection of physics and AI, allowing me to scale my ideas and deepen my impact.

How has your experience as a teaching assistant and research mentor shaped your academic development?
Serving as a teaching assistant and research mentor has been one of the most rewarding parts of my academic journey. Teaching physics to undergraduates has deepened my own understanding of core principles, while also strengthening my communication and leadership skills. Breaking down complex topics into simple, relatable ideas has made me a clearer thinker and a better researcher.

Mentoring students in research has also helped me grow. Guiding them through problem-solving, debugging code, or interpreting simulation results pushes me to stay sharp and up-to-date. It’s also fulfilling to watch them progress—from being unsure to confidently presenting their work. These roles have not only shaped me academically but have also reinforced my passion for collaborative learning and knowledge-sharing.

Looking ahead, how do you see your work influencing real-world applications in physics, materials science, or data-driven technologies?
My research on compressible microgels is contributing to a relatively new frontier in soft matter physics. These materials can deform and adapt to their surroundings, making them ideal candidates for smart and responsive systems. In fact, researchers have already begun exploring their use in building adaptive structures and actuators.

By modeling the behavior of compressible microgels from a fundamental perspective, my work helps lay the theoretical groundwork needed to design materials that can be used in applications like drug delivery, biomedical devices, soft robotics, and tissue engineering.
On the data-driven side, my machine learning projects—such as fisheye object detection—showcase how AI can be applied to challenging real-world problems involving distorted imagery, which has implications in autonomous vehicles, surveillance, and industrial robotics.

Ultimately, I see my work as bridging theory and practical innovation, contributing to the next generation of intelligent, adaptable materials and systems.

What advice would you give young Africans or international students interested in pursuing a PhD in the U.S. or working at the intersection of physics and AI?
First, believe in your potential—don’t let where you’re coming from define where you’re going. Many of us from Africa often face limited resources, but resourcefulness, consistency, and vision can take you far. Be intentional about learning, especially the fundamentals of physics, math, and programming—they’re the building blocks for solving complex problems at the intersection of physics and AI.

Second, find your niche. It’s a big field, so focus on a specific problem you care about and go deep. For me, it was the behavior of compressible microgels, and that opened doors to both physics and AI innovation.

Third, don’t do it alone. Seek mentors, communities, and collaborations. Your journey will be easier if you surround yourself with people who challenge and encourage you. And finally—apply! Many opportunities are just one bold step away.