Meet Connor Hill, the 17-year-old Pennsylvania student who wrote a computer program to classify noble polyhedra; he proved there are 146 isolated examples and won $250,000
By writing a custom computer algorithm that mapped every possible version of a rare class of geometric shapes, 17-year-old Connor Hill from Port Matilda, Pennsylvania, has won the top $250,000 prize at the 2026 Regeneron Science Talent Search.His computational proof classified a complex group of three-dimensional shapes called noble polyhedra. His work identified two infinite structural families and exactly 146 separate, isolated examples.The announcement was made jointly by Regeneron Pharmaceuticals and Society for Science at a black-tie ceremony at the National Building Museum in Washington, D.C. The competition, now in its 85th year, is widely recognised as the oldest and one of the most prestigious science and mathematics competitions for high school seniors in the United States.
Solving a long-standing geometry problem
Polyhedra are three-dimensional shapes made from flat polygonal faces and straight edges. Common examples include cubes, pyramids and octahedrons.A polyhedron is called noble when it has a high level of symmetry. Every face must have the same shape, and every vertex, or corner, must have the same arrangement of faces. A cube is a simple example because all six faces are squares and all eight corners have the same structure.Noble polyhedra, however, allow more complicated and even self-intersecting shapes than the familiar Platonic solids. This makes their classification a much harder mathematical problem.Before Hill’s research, mathematicians had already identified two infinite families of noble polyhedra. However, the complete list of individual, isolated shapes was still unknown.By 2020, researchers had confirmed 61 isolated examples, but they believed that many more could exist because of more complicated combinations of symmetry.To solve the problem, Hill created a custom computer program that could systematically examine all possible geometric arrangements. Instead of trying to search through an endless number of three-dimensional shapes directly, he converted the problem into algebraic equations, including polynomial equations. This allowed the computer to reduce the problem to a finite set of possibilities that could be checked.According to research summaries released in the Society for Science 2026 Regeneron STS media kit, Hill’s program completed the exhaustive search and showed that there are exactly 146 isolated noble polyhedra, in addition to the two known infinite families.
Connecting pure mathematics with other areas of science
Hill’s work was not only about finding the number of noble polyhedra. He also believes that the method he developed could be useful in other areas of science and mathematics.In a discussion published on Regeneron’s scientific portal after the award announcement, Hill spoke about his approach with Dr. Henry Wei, Executive Director of Development Innovation at Regeneron.Explaining how he solved the problem, Hill said:“By translating the problem into a different mathematical framework, I could narrow it down to a finite set that could actually be solved.”Hill said his interest in the problem began through his involvement in an online geometry community, where the classification of noble polyhedra had been discussed as an open mathematical challenge.He also explained why mathematics is useful for solving difficult scientific problems:“Science gives us the concepts, but math allows us to formalize them and apply them. When we’re solving big problems, the key is breaking them into smaller pieces that math can address. Often, breakthroughs happen when someone connects ideas across disciplines.”Dr. Wei pointed out that Hill’s approach of turning a complicated problem into smaller, solvable parts is similar to methods used in modern biomedical research. Scientists use mathematical models to study problems such as protein folding, disease development and how drugs interact with the body.Hill hopes his computational method can eventually be used to study compound polyhedral structures and connected geometric networks.
National awards and support for young researchers
The 2026 Regeneron Science Talent Search awarded more than $1.8 million to 40 high school finalists. The finalists were selected from thousands of students who applied from across the United States.Second place and a $175,000 prize went to 17-year-old Edward Kang of Hackensack, New Jersey. Kang developed an artificial intelligence screening tool called RetinaMind.Kang trained machine-learning models using retinal scans to identify subtle blood-vessel patterns associated with autism and attention-deficit/hyperactivity disorder. He also created cellular models to study related genetic changes.Third place and a $150,000 prize went to 17-year-old Iris Shen of The Woodlands, Texas. Shen studied possible treatments for blood cancer using clams as a testing model. She observed cellular responses that were similar to those seen in human drug trials.Maya Ajmera, President and CEO of Society for Science and Executive Publisher of Science News, praised the finalists in an official statement released by Society for Science.“Congratulations to the winners of this year’s Regeneron Science Talent Search,” Ajmera said. “Their bold vision and perseverance reveal what the next generation of problem solvers truly looks like—and why our future is in capable hands. Their creativity, ambition and courage to confront the world’s toughest challenges are exactly what this moment demands.”Regeneron also highlighted the importance of supporting young researchers. Dr. George D. Yancopoulos, co-founder, Board Co-Chairman, President and Chief Scientific Officer of Regeneron, is himself a 1976 winner of the competition.“Congratulations to the winners of the 2026 Regeneron Science Talent Search, and to all the finalists who participated in this year’s competition,” Yancopoulos said. “These students represent exactly the kind of extraordinary talent scientific progress depends on.”Alongside the 2026 awards, Regeneron announced a 10-year extension of its title sponsorship of the competition through 2036. The company pledged another $150 million to support STEM education programs and scientific competitions run by Society for Science.Hill plans to attend the Massachusetts Institute of Technology (MIT) in the autumn, where he will continue studying computational mathematics.
