NCSA and NVIDIA teamed up in this year’s Big Q Hackathon, which took place in Chicago and is hosted by the Chicago Quantum Exchange in partnership with QuantX. The competition lasted four days, with different phases during the first two days and the last two days. This is the first time the Hackathon was held in the U.S. – previous Hackathons were held in Europe and Canada.

NCSA was the only non-profit organization competing – the Center’s resources went head to head with giants like AWS, Quantinuum and IBM. That makes their showing all the more impressive, as they were pitted against corporate giants with the resources to outmatch most academic institutions. The first phase, called the Technical Phase, was days one and two of the competition. Nine technical teams competed during this phase, and each used computing platforms from the competing institutions. NCSA resources were paired with NVIDIA resources for both phases of the competition.

Bruno Abreu, NCSA/IQUIST research scientist and quantum computing co-lead, was excited to have the opportunity to showcase what NCSA resources could do. “Our Center has a well-structured quantum computing strategy,” he said, “and this Hackathon showed that we are very intentional about deploying our best efforts to materialize quantum advantage soon with an application-oriented approach, offering research consulting and advanced cyberinfrastructure. In partnership with IQUIST, we are working hard to consolidate Illinois as a leader in quantum technologies, and we have many exciting opportunities coming soon!”

The teams assigned to NCSA’s resources were specifically chosen from universities other than the University of Illinois Urbana Champaign to ensure that no team had a homecourt advantage. Mentors for the teams assigned to NCSA resources were Bruno Abreu, Pooja Rao, quantum application engineer from NVIDIA, and Joel Thompson, associate technical fellow from Boeing.

The NCSA/NVIDIA teams were given a challenge from Boeing to solve, and they were required to use NCSA resources paired with the CUDA Quantum platform to find the best solution to the challenge. “CUDA Quantum,” as NVIDIA’s Rao explained, “is NVIDIA’s open-source platform for hybrid quantum-classical computing, built for high-performance, scalability, and ease-of-use. As all valuable quantum applications of the future will be hybrid, CUDA Quantum enables users to develop performant hybrid applications that can easily scale to supercomputing scale systems like NCSA’s Delta.”

During the Technical Phase of the competition, the team assigned to NCSA’s resources named themselves the Deep Dish Qubits Team. Team members included Rebecca D’Antonio, Jason Shao and Angela Abongwa from the University of Chicago (UChicago) and Saarah Nazar from Purdue University. 

D’Antonio is a master’s student at UChicago’s Pritzker School of Molecular Engineering, where she’s taking a computational materials modeling concentration. Prior to enrolling in graduate school at UChicago, D’Antonio received her bachelor’s in computer science from Louisiana State University. “I am finishing my last quarter currently,” she said, “after which I am hoping to begin working in industry with quantum computing.”

Abongwa is a third-year undergraduate student at UChicago, majoring in physics with an interest in quantum physics. “I recently joined one of the Quantum engineering labs at my college,” she said. “I am still exploring the different areas of research in the lab, but its primary focus is spin dynamics.”

The students found themselves a little intimidated at first, especially since they were using platforms they were unfamiliar with. “Prior to the Hackathon, I only had experience with IBM’s Qiskit software to build quantum circuits, so I was really glad I got the chance to work with NVIDIA’s CUDA Quantum software,” said Abongwa. For D’Antonio, this was her first such competition.”The Big Q Hackathon was actually my first hackathon experience, so I had no idea what to expect going into it. I found myself very nervous about how I would be able to contribute and whether my mentors and peers would be willing to help and extend their knowledge.”

The team found they had the support to get them up to speed quickly. “Throughout the entire process, [mentors] Bruno, Pooja, and Joel remained available and helped us to navigate issues,” said D’Antonio. “They were extremely patient with our questions, and they made sure to explain things to us in a way that would not only help us understand the technical problem we were facing but also how to extend our knowledge to tackle other quantum computing challenges going forward. They guided us while also encouraging us to do our own critical thinking and critical analysis. By the end of the weekend, we all felt accomplished and much more knowledgeable in applying quantum algorithms to real-world use cases.”

With a powerful team-up of talented students and attentive mentors, there was still one unknown factor for the competitors. Recall they were chosen specifically because of their unfamiliarity with NCSA and NVIDIA resources. They were in luck; NCSA has prided itself on its efforts to make Center resources easy to use for all researchers, no matter their background.

“The NCSA resources were well documented and easy to use,” said D’Antonio. “They proved to be extremely helpful in accelerating our calculations, as we were able to run very large systems – 30 qubits! – in just a few minutes, which is a massive feat in quantum computing. Without access to these resources, we would not have been able to generate significant and substantial results in such a short period of time. I work with HPC resources for the Gagliardi Group and found NCSA’s to be comparatively digestible to navigate.”

D’Antonio described the challenge they were presented with during the first phase of the hackathon. “Three other students and I worked with mentors Bruno A, Pooja R (Nvidia), and Joel T (Boeing) to generate an optimization solution to a ply-composite model for Boeing. We used CUDA Quantum and NCSA’s Delta supercomputer to run these calculations and put together a presentation of our results.” These results would play an important role in the second phase of the hackathon when the business teams took over, but to get to that point, the students had to work as a team to get the best results possible before handing off the baton.

“We spent the first few hours of the challenge simply breaking down the proposal as a group and offering ideas on how best to tackle the problem,” said D’Antonio. “We then worked together to figure out how to adapt our approach with a CUDA Quantum implementation, which was a quantum platform that none of us were familiar with, as well as how to run our code on the NCSA’s computing resources. From there, we split into separate tasks and began a style of pair-programming, where two people worked on the algorithm, and the others worked to either generate and tabulate results or come up with test cases.”

The Deep Dish Qubits Team took third place during the first phase, an incredible feat considering they were up against teams that used resources from huge companies. Their results would prove to be instrumental in the second phase of the competition, allowing the business team to compete at the highest level.

The students took away more than their award for third place. D’Antonio couldn’t speak more highly of the experience. “I was astounded by how generous and encouraging every person I met was at the hackathon. It was a privilege to connect with so many brilliant people who were more than willing to talk about their areas of expertise and to learn about the many budding applications of quantum computing to real-world issues in the near term. I felt empowered and excited about the future of quantum, as well as much more equipped to apply what I have learned in a classroom to these cases. I have also learned the power of working with a team of people who communicate clearly and remain respectful and encouraging at all times. I would recommend this experience to anyone who is able, and once again am grateful to the mentors and team members whom I had the pleasure to work with.”

The ply-composite optimization problem that the Deep Dish Qubits team developed for Boeing was a fantastic use-case enabled by CUDA Quantum, demonstrating what can be achieved towards industry problems using the best combination of GPUs and emulated quantum systems.

–Pooja Rao, quantum application engineer, NVIDIA

The second phase of the Hackathon ran differently than the first. Teams were again comprised of students who weren’t affiliated with the resources they based their pitches on to help keep a level playing field. For this phase, the Boeing/NVIDIA/NCSA team was made up of three students: Michael Gustafson and Pablo Matarredona from the Chicago Booth School of Business, and Ariadna Fernandez, a recent graduate from UIC. 

Fernandez received her bachelor’s in computer science from UIC earlier this year. She’s currently working as a consultant. “I have been exploring the field of quantum information science and technology since the summer of 2022 when I had the opportunity to participate in the first cohort of the Open Quantum Initiative Fellowship by the Chicago Quantum Exchange, which also co-organized the hackathon,” she said.

Matarredona is currently an exchange student at the Booth School of Business. His home university is the Rotterdam School of Management at Erasmus University in the Netherlands. “I’m studying Management of Innovation and Technology, and I have a background in engineering,” he said. “I did my master’s thesis on the emergence of the quantum computing industry and its impact on the energy transition. During the summer, I worked at Applied Materials, a semiconductor equipment company, where I also did some work on quantum computing and its strategic implications for the semiconductor industry.”

The students named their team Teq Staq and were given the solutions that the Deep Dish Qubits came up with to start their phase of the competition. The teams had to take these results and turn them into the most compelling pitch to beat out the competition. 

“We were given one of the top solutions from the technical phase and were tasked with coming up with a pitch that would convince the original company, in our case Boeing, of the importance of using quantum technology to solve this specific problem,” said Fernandez. “The specific industry problem was finding the best way to stack ply composites used in building airplanes. We pitched the solution the technical team came up with and expanded on its scalability and potential value to the company and industry at large.”

The winning approach would require Teq Staq to brainstorm and connect as team members fast. “Because of the short time we had,” Fernandez explained, “we quickly jumped into understanding the problem that Boeing posed, why it was important to them and why a quantum lens is necessary.”

“Once the technical reasoning was outlined,” Matarredona said, “we tried to break down the unit economics and see the economic impact of such a solution. Other points of view, such as environmental and operational efficiency, were taken into account to come up with a comprehensive and sound business case.”

“[Our] team then conducted research on the market and how much it would impact not just Boeing but the entire industry if this problem were to be solved,” said Fernandez. “The technical team was able to run their algorithm in NCSA’s Delta system using NVIDIA’s CUDA hybrid simulation platform, and it yielded promising results that helped elevate our pitch about the applicability and scalability in the short-term.”

Both Fernandez and Matarredona shared the sentiment that the Hackathon was more than a worthwhile experience. “This was my first quantum hackathon, and I had the best time,” Fernandez said. “I really enjoyed every minute of it! I learned so much and was enlightened by the conversations with other participants and mentors. Being part of the business team was a great learning experience for me and gave me a different perspective about what is important when we think of short-term quantum computing applications and the business considerations. I learned this especially from team members, Pablo and Mike.”

“I learned a lot,” Matarredona said. “My biggest takeaway is that quantum approaches for computation are much closer than I would have expected, especially quantum-inspired approaches. Also, I found that these sorts of experimental working environments allow for innovation to flourish.”

At the end of their phase, all their hard work and collaboration paid off, with the solutions provided by the Deep Dish Qubits helping to push the business team into first place, an incredible feat considering the NCSA and NVIDIA resources they were pitching were being compared with some of the biggest names in quantum computing.

NCSA mentor Abreu was thrilled with the win. He had words of high praise for the teams’ fantastic results. “We had a unique combination of resources, and the teams brilliantly managed to take the most out of them,” he said. “Boeing provided us with an exciting use case, and seeing their progress toward such a complex solution in such a short time, both in the technical and business phases, was imposing.” 

The team was ecstatic about their win, and both members interviewed shared their win with the Technical team. “Winning felt amazing,” said Matarredona. “A great validation of our work! It was great to see that unlocking quantum value is not as far as people would expect. Kudos to the technical team. They did a great job developing a useful solution. As part of the business team, our work was much smoother thanks to them.”

“It was really exciting!” said Fernandez. “We couldn’t have done it without our amazing mentors and, of course, the technical team who did a fantastic job creating and running the algorithm with promising results. That was the perfect combination that set us up for success.”