It is widely accepted that the experiments related to the LHC at CERN have taught us all they are able to about the world - at least at the current intensity levels of the particle collisions (the Luminosity in collider physics terms). So the name of the game is "turn it to eleven". But with higher luminosity follows more collisions and with more collisions follow more data. Ultimately, the current conventional algorithms for tracking, calibration, analysis and everything in between will not be fast enough to keep up with the ever-increasing flow of data. Newer methods - preferably GPU-based - are needed, and that's where Machine Learning comes in. Being constructed solely from matrix operations, ML algorithms can be run on GPUs making them fast - really fast. 

My research concerns implementing these cutting-edge methods within the ALICE experiment to enable real time calibration and processing of data. Additionally, some ML algorithms have the potential to surpass analytical methods in performance. For the envisioned ESSnuSB experiment, I'm investigating if GNNs can surpass the current tools when it comes to event classification and reconstruction resolution.