New Delhi: An 18 year old American student Matteo Paz used AI to survey 200 terabytes of all-sky infrared data captured by NASA’s WISE telescope after
a decade. This represented 200 billion lines of data. Paz developed waveform-based machine learning methods to sort the entire catalogue, and detect as well as characterise objects that were changing their brightness in infrared wavelengths. The search yielded a census of 1.9 million objects that vary in infrared, 1.5 million of these are new discoveries. NASA Administrator Jared Isaacman offered to hire Paz as a scientist at NASA.
Where do I sign? pic.twitter.com/iV49Oagn37
— Matteo Paz (@matteopaz06) December 27, 2025
Paz conducted the study as a staff researcher under NASA funding. The new discoveries by Paz include a wide range of objects that change their brightness over time, including pulsating stars, distant supermassive black holes that are actively feeding on the surrounding gas and dust, embryonic protostars being born within stellar nurseries, multiple star systems, stars dying violent deaths as supernovae, and stars that are particularly active. The goal of the project was to catalogue infrared variable objects, and calculate parameters for use in further research.
Potential for future research
The research has uncovered a number of Cepheid Variable stars, supernovae and old, low-mass pulsating stars known as RR Lyrae stars, that all act as standard candles and help resolve the Hubble Tension, an ongoing crisis in cosmology on the rate at which the universe is expanding. The search has also uncovered a number of distant galaxies. A paper describing the research has been published in the Astronomical Journal. Paz attends Pasadena High School where he is the president and founder of the research club, where he mentors other students in science contests.
