Astroparticle physics studies high-energy particles with cosmic origins that are detected on Earth. Emerging more than a century ago with the detection of cosmic rays, this field now deals with high-energy gamma-rays and neutrinos. Over the last few decades, astroparticle physics has evolved towards multi-messenger astronomy, where information from multiple messengers such as photons, gravitational waves, and cosmic particles is combined to gain a deeper understanding of astrophysical phenomena. The development of new instruments and connecting different telescopes into a common network is bringing the field into the petascale era, where data science techniques are becoming crucial for future discoveries. In this talk, I will give an overview of astroparticle physics and multi-messenger astronomy, highlight the latest achievements, and discuss the synergies between astrophysics and data science, with examples from the Astroparticle Physics Lab at JetBrains Research.

Multi-messenger astronomy has gained significant interest in recent years, resulting in a substantial increase in published observation reports. Human-written reports such as GCN Circulars and ATels provide important interpretations and discussions of observations, but lack a defined format, making it difficult to analyze and cross-link data to enable the discovery of new patterns.

In the field of natural language processing, named entity extraction (NER) refers to the process of identifying named entities such as people, organizations, and other relevant terms in a particular domain. Developing an end-to-end NER system that harnesses the power of large language models is our approach to overcoming the challenges typically associated with NER, including limited labeled datasets and textual ambiguity.