Farewell to the 2024-2025 Frontera Fellows: Bold Journeys, Bright Futures

TACC is bidding farewell—for now—to the 2024-2025 Frontera Computational Science Fellows. Over the past year, these exceptional graduates have leveraged the power of the Frontera supercomputer and collaborated with leading experts at TACC, pushing the boundaries of computational science.

Read about the impact of their work, what they learned, and why they believe future students should apply to this unique opportunity of the Frontera Fellowship.

Ayodeji Omoniyi

Academic Institution: Stevens Institute of Technology

Field of Research: Chemical Engineering (Computational Heterogeneous Catalysis)

What were the most impactful project(s) you worked on as a Fellow, and what role did you play in them?

I advanced my Ph.D. project on the design of bimetallic catalysts for upgrading carboxylic acids by using the AI-focused Vista cluster to perform required ground- and transition-state DFT calculations to map out reaction pathways and guide catalyst selection. I also conducted a technoeconomic analysis evaluating the feasibility of establishing a biodiesel production plant in the Northeastern United States. Finally, I contributed to the development of machine-learned interatomic potentials for investigating reconstruction in bimetallic catalysts, which involved running thousands of ground-state and ab initio molecular dynamics DFT simulations.

From a scientific, computational, or professional lens, what were the most valuable lessons you gained?

At TACC, the standard of excellence shows how much smoother and more impactful scientific progress can be if each part of the system performs at a high level. The competencies I observed during the Fellowship were outstanding. Furthermore, I have a renewed appreciation for supercomputer architecture details. Before the Fellowship, I focused mostly on the computational chemistry and chemical engineering aspects of my research—with limited attention to specifications of the supercomputers that I used to run my calculations. Thanks to this experience, I pay closer attention to the architecture and performance intricacies that make my supercomputer simulations possible.

What did you enjoy most during your time as a Fellow?

I enjoyed the incredible treatment, reception, and support I received from everyone at TACC. When describing the experience to my friends, I often used the phrase “celebrity treatment.” TACC staff consistently prioritized our needs, helped solve our problems, addressed our inquiries, and showed genuine care for our well-being, progress, and overall research.

How has the Frontera Fellowship prepared you to advance in your research field?

The Fellowship has helped me stand out in any pool of applicants, opening doors for my research to gain visibility and serving as a catalyst (pun intended, considering my research) for pushing my work to the front of the line. Thanks to this experience, I have won awards and recognitions. The Fellowship has helped and will continue to help my work get a foot in the door, and I am grateful for this distinguishment.

What guidance would you offer to new Fellows, and what makes this program a valuable opportunity for others?

The Frontera Fellowship is an extraordinary honor. One of the most valuable aspects is the unparalleled access to TACC’s family of supercomputing clusters. For any researcher relying on computational resources, these clusters represent incredible research-advancing tools with their speed, ease-of-use, and a dedicated team of experts available to help with troubleshooting. To future Fellows: Fully utilize these resources and do not hesitate to work with TACC staff to elevate your research.

Chishan Zhang

Academic Institution: University of Illinois Urbana-Champaign

Field of Research: Ph.D. candidate, Geography & Geographic Information Science

What were the most impactful project(s) you worked on as a Fellow, and what role did you play in them?

Using Frontera, I designed and implemented a scalable workflow to process massive satellite datasets. I also integrated process-based crop models and utilized Model-Agnostic Meta-Learning for robust cross-regional soybean yield predictions in the Americas. A key part of my work involved enhancing nearly real-time predictions by incorporating climate pattern teleconnections like the Madden-Julian Oscillation.

From a scientific, computational, or professional lens, what were the most valuable lessons you gained?

I learned to tailor modeling approaches to HPC architectures like Frontera, such as how to design and optimize my approach to processing vast amounts of data and run complex simulations at an unprecedented scale. I gained a deeper understanding of scaling geospatial analysis and the importance of addressing regional diversity and climate variability in large-scale agricultural modeling.

What did you enjoy most during your time as a Fellow?

I enjoyed engaging with other Fellows and the TACC community, learning how HPC is applied across diverse disciplines. I loved hearing inspiring stories about how each Fellow chose a career in science. Gaining new career development suggestions was nriching.

How has the Frontera Fellowship prepared you to advance in your research field?

This experience has given me the confidence to tackle large-scale analysis, which was instrumental in securing a postdoctoral fellowship where I will be conducting global-scale research. The research I conducted during the Fellowship has led to one publication currently under review and another actively in preparation.

What guidance would you offer to new Fellows, and what makes this program a valuable opportunity for others?

I would advise new Fellows to fully leverage Frontera’s unique scale and never hesitate to consult TACC’s experts because their support is invaluable. This program empowers you to tackle grand scientific challenges and gain essential HPC skills critical for cutting-edge research. Just apply—the experience is transformative!

Mikayla Darrows

Academic Institution: University of Florida

Field of Research: Ph.D. candidate, Chemistry

What were the most impactful project(s) you worked on as a Fellow, and what role did you play in them?

I focused on improving the computational efficiency of LEGOLAS, a machine learning model that predicts protein chemical shifts. TACC’s supercomputing resources allowed me to adapt the model to run on the new AI-focused Vista Grace-Hopper GPU nodes and parallelize its inference process across a large-scale, multi-GPU setup. These improvements enabled me to generate chemical shift predictions across hundreds of thousands of protein trajectories, creating a large-scale dataset to serve as a resource for connecting Nuclear Magnetic Resonance (NMR) chemical shifts to protein structures in chemistry and biology research.

From a scientific, computational, or professional lens, what were the most valuable lessons you gained?

I gained a deeper understanding of HPC and the infrastructure behind large-scale machine learning. Throughout the Fellowship, I took ownership of adapting models to run efficiently on specific GPU architectures. This experience pushed me to think like a computer scientist—not just a chemist—and helped me appreciate how performance tuning and hardware-aware programming impact scientific progress.

What did you enjoy most during your time as a Fellow?

I enjoyed connecting with scientists outside of my discipline. The Fellowship opened my eyes to the wide range of fields that use similar computational resources and broadened my perspective on what’s possible with HPC. Although I am a chemist at heart, I valued talks about hardware design and system architecture, which gave me a newfound appreciation for the technology that makes my research possible.

How has the Frontera Fellowship prepared you to advance in your research field?

TACC’s computing resources and the guidance of mentors, staff, and support teams allowed me to push the boundaries of what my machine learning model could do, reaching areas of research I might have otherwise overlooked.

What guidance would you offer to new Fellows, and what makes this program a valuable opportunity for others?

One of the most valuable aspects of the program was the opportunity to take initiative and guide my own research path. That level of independence can be daunting, but with the right support, it is incredibly rewarding. Therefore, take full advantage of the TACC community. Attend events, ask questions, and connect with staff. The environment is incredibly supportive and full of people who are invested in your success.

Jeffrey Lotthammer

Academic Institution: Washington University, St. Louis

Field of Research: Ph.D. candidate, Computational Biology, Biophysics, Bioinformatics

What were the most impactful project(s) you worked on as a Fellow, and what role did you play in them?

I am excited about my ongoing work combining deep learning with disordered protein sequence design with applications spanning basic mechanistic science, biosynthetic technologies, and translational cancer research. I am interested in understanding the biology of intrinsically disordered proteins, which—unlike folded proteins—lack a well-defined 3D structure and often are poorly conserved by amino acid sequence while still carrying out many essential cellular functions. I have been developing tools to explore how nature encodes robust molecular function in these proteins despite the absence of conserved structure or sequence.

From a scientific, computational, or professional lens, what were the most valuable lessons you gained?

I learned how to scale code efficiently across large distributed systems while deepening my understanding of parallel computing basics. This experience helped me figure out when and why you need to make certain design choices or trade-offs to keep workloads balanced and everything running smoothly. This was especially useful when scaling my deep-learning workloads across tons of networked nodes, where I gained hands-on experience debugging communication and compute bottlenecks. I now understand how different design decisions can shift that balance.

What did you enjoy most during your time as a Fellow?

I loved advancing my research and meeting so many amazing people. From the scientists at TACC to the other Fellows—as well as the supportive business staff and administrators who make programs like this possible—I’m grateful for TACC’s support structure and the connections I made.

How has the Frontera Fellowship prepared you to advance in your research field?

Having access to TACC’s HPC resources allowed me to pursue scientific research that I likely wouldn’t have been able to tackle otherwise. This Fellowship has significantly equipped me to handle even larger computational problems with confidence.

What guidance would you offer to new Fellows, and what makes this program a valuable opportunity for others?

This is an incredible opportunity that gives you the freedom and flexibility to dive into the science that excites you most. Don’t be afraid to ask for guidance or support. Even if someone can’t directly help, the mentors are happy to connect you with the right person to keep your research moving forward.