The UK Quantum Hackathon, organized by the National Quantum Computing Centre (NQCC), has become an annual highlight for the UK’s quantum community. Enabling students and early-career researchers to work alongside industry mentors and technology experts, the three-day event provides an immersive learning experience that tests the capabilities of current quantum computers whiles also hinting at the prospects for their future deployment.
In the 2026 event, the fifth in the series, ten teams of hackers gathered at the University of Warwick to tackle an eclectic mix of problems set by mentors from different end-user organizations. During the hackathon they had just two days to explore the use case, formulate a model solution, and generate results using a range of quantum emulators and physical processors. “Our industry mentors join the teams to provide specific domain knowledge and contextualize the use case, while the teams are also supported by our own applications engineers and technology providers such as IBM and Amazon Web Services BraKet,” says the NQCC’s Daisy Shearer, who took the lead role in organizing the hackathon.
The ten use cases selected for this year’s event showcased problems that could nurture business growth while also generating tangible benefits for people and communities. Those dual objectives were exemplified by the problem set by Nexorion Spacetech, a start-up company formed in 2025 that is focused on developing hybrid quantum–classical solutions to optimize the positioning and deployment of individual satellites within a larger constellation.
Such constellations of satellites are often used for Earth observation, with commercial operators using scheduling software to determine which satellite should perform each imaging task. Factors to consider include the urgency of the request – capturing the leading edge of a wildfire, for example, would take priority over routine observations of crop health – along with the location and trajectory of each satellite relative to the target. “This is a hard combinatorial problem,” said Utsav Talekar, one of the company’s co-founders. “Exact solvers cannot return answers within the available planning window for large fleets containing dozens of satellites, so operators must rely on approximations that provide a workable solution rather than the best one.”
Nexorion challenged its team to devise a quantum algorithm that would generate a useful starting point for a classical optimization program. Such “warm-start” solutions are widely used to help combinatorial solvers to converge on a result more rapidly, and the aim was to compare the performance of a quantum-powered warm start with its classical equivalent. For the quantum approach the hackers used an algorithm that examines many possible solutions at the same time, picking the ones that are most likely to yield the best result.
When run on IBM Quantum’s Heron device, they found that the quantum algorithm performed as well as the exact solver. However, the team could not beat the classical solution for the problem size that could be tackled in the hackathon, which was limited to eight satellites. “What was really important was the way that they improved the quantum circuit to achieve a better outcome,” said fellow co-founder Thien Bui. “They really engaged with the larger scale problem, not just what can be achieved in two days.”
Making a difference
Reflecting the core ethos of the hackathon to consider the wider impact of quantum computing, the team analysed the effect that improved satellite scheduling could have on key stakeholders. They noted that more timely imaging could enable emergency responders to receive information about natural disasters or extreme weather events as they unfold, allowing them to target their resources and provide early warnings to affected communities. But the team also recognized public concerns about such co-ordinated satellite imaging being used for surveillance, suggesting that acceptable-use policies and clear regulation should be established from the outset. “It was clear that they had really thought about the broader implications of their use case,” said Natasha Oughton, who leads the NQCC’s initiative for responsible and ethical quantum computing.
For Nexorion, the hackathon provided valuable input into the development of its early-stage applications technology, while also allowing Bui and Talekar to forge new connections with hardware providers and the NQCC. Their long-term goal is to develop a hybrid quantum solution for co-ordinating a satellite imaging system that could guide future landings on the Moon or Mars. “We have been selected by the European Space Agency to validate the concept of using hybrid quantum computing for these landing systems, which need real-time feedback to guarantee the safety of the spacecraft and the astronauts,” said Talekar. “It will be a long journey, but being part of the hackathon has given us a real boost.”
Applied Quantum Computing (AQP) is another applications-focused company that works with its customers to explore practical use cases that can benefit from near-term quantum computing. It has particular expertise in the healthcare sector, and for the hackathon the company challenged its team to test the capabilities of quantum machine learning for predicting the onset of brain seizures from the electrical activity recorded by an EEG (electroencephalogram). “A reliable predictor could offer an early warning that someone is at risk from a seizure, providing timely access to support and enabling more effective clinical care,” said AQP’s Ethan Davies, who was mentoring the team.
Electrical clues
Previous studies have linked subtle changes in electrical activity to a pre-seizure transition state in the brain, and for this challenge the hackers were given 10-minute data streams derived from an EEG that were labelled as “normal” or “pre-seizure”. To make the problem more tractable, Davies had already condensed the raw EEG data into a series of relevant features that could be used as the input for both classical and quantum models. But further consolidation was still needed to create a feature set that was small enough to be processed with the available resources. “It was really interesting to see the strategies the team used to select the features that would make a meaningful difference to the predictive capability of the model,” said Davies.
Initial simulations of a quantum neural network suggested that it would not be able to match the performance of a classical model, although a last-minute run on a real quantum processor produced better results than had been predicted. The team also tried a hybrid approach, in which a so-called quantum kernel was used to create a richer feature space that could be used as the input for classical machine learning.
The quantum kernel method seemed to produce better outcomes, and in a practical setting would avoid the need to make constant calls to a quantum computer. It was still unable to outperform a classical solution, however, but the team suggested that the results were promising enough to warrant further investigation with more diverse datasets. Davies also believes that the findings could be applied to other use cases in the healthcare sector. “Quantum methods like these could help to predict the outcomes of sudden events, such as the immediate effect of taking a patient off a mechanical ventilator,” he said.
After all the final presentations, AQP’s team won the glory of the Best Team Name competition with their chosen moniker of “404: Medical Student Not Found”. Meanwhile, the overall winners had tackled another use case that combined a business opportunity with clear benefits for the wider world. Mentored by Tech Mahindra, a multinational company specializing in information technology, the goal was to identify mobile transmitters that could be switched off during periods of low demand. Accurately predicting which transmitters could be put to sleep without affecting users would allow mobile operators to save money, while also reducing energy consumption and carbon emissions.
By simulating a street zone in Kensington, the team found that an existing classical solution could reduce cell usage by up to 40%. At this scale the classical approach performed better than the quantum solution, but the team pointed out that the computing resources needed for the classical solver would increase exponentially as the simulated region is extended to metro areas. In contrast, the resources needed for the quantum algorithm would scale much less rapidly, either quadratically or even linearly.
Intriguingly, the second-placed team tackled the first musical problem to be set at the Quantum Hackathon. They investigated whether a quantum solution could accelerate efforts to uncover related melodies in Irish folk music, with the aim of helping researchers to identify families of similar songs from an archive of more than 200,000 records. While the quantum method achieved some success, the team concluded that human musicians still have the best ear for picking out a tune.
“It’s been wonderful to hear about all these fascinating use cases, which are way beyond anything we might have imagined even 20 years ago,” said Gerard Milburn, one of the judges and both Quantum Fellow of the NQCC and a professor at the University of Sussex. “I am particularly thrilled to see a use case for Irish folk music, it’s so unexpected and offers a hint of how quantum computing will have an impact beyond science and technology to address wider cultural perspectives.”
The post From space to folk music: quantum hackathon explores diverse use cases appeared first on Physics World.