Supercomputing Hackathon to bring students and specialists together to tackle the challenges of data-driven astronomy

The initiative, organised by CUDI and LAMOD with infrastructure support from PIG and Grid-UNAM, will take place on 22 and 23 August in a hybrid format and will bring together participants with varying levels of experience to develop data science solutions for the classification of variable astronomical objects.

Contemporary astronomy is undergoing an unprecedented transformation due to the exponential growth in the volume and complexity of data generated by large observatories and astronomical surveys. In response to this challenge, Mexico’s National Research and Education Network, CUDI, has announced the Hackathon: Time Series Classification with Applications to Astronomical Data, an event that will enable students, researchers and professionals to apply data science and machine learning techniques to real-world problems in modern astronomy.

The event will take place in a hybrid format, with an in-person component at the Institute of Astronomy of the National Autonomous University of Mexico (UNAM), on 22 and 23 August, and aims to promote the use of advanced computing infrastructure for the analysis of scientific data.

The main objective of the hackathon is for participants to design, implement and evaluate computational models capable of classifying astronomical time series, working with datasets from major international projects. In addition to the accuracy of the models, the organisers will assess aspects such as the reproducibility of the analyses, the scientific interpretation of the results and the ability to clearly communicate the methodologies used.

To facilitate participation by people with varying levels of experience, the competition will be structured into three categories. The basic level will focus on the classification of light curves from the OGLE project, allowing teams to familiarise themselves with fundamental techniques for analysing astronomical data. The intermediate level will use data from Gaia, incorporating additional photometric and astrometric information to enrich the classification models. Finally, the advanced category will tackle scenarios inspired by Rubin/LSST, characterised by large volumes of data, incomplete information and challenges associated with the detection of transient phenomena.

The call for participants is open to secondary and higher education students, as well as researchers, professionals from public or private organisations, and data science enthusiasts. Although basic knowledge of programming and data analysis is required, no prior experience in astronomy is necessary, as the fundamental concepts related to the challenge will be introduced before the activities begin.

According to the organisers, this initiative represents an opportunity for participants to get involved in scientific projects supported by high-level computing infrastructure and to work with data generated by real astronomical observations. Furthermore, the hackathon aims to strengthen skills in analysing large volumes of data, machine learning and advanced computing – competencies that are increasingly relevant to research and innovation across multiple disciplines.

Through activities such as this, the region’s academic and technological community continues to promote the development of specialised talent and the use of advanced infrastructure to tackle the scientific challenges of the data age.

Further information and registration: CUDI, Hackathon: Time Series Classification with Applications to Astronomical Data.