I am a Data Scientist working at the intersection of AI and environmental science to tackle real-world challenges with data-driven solutions. At WeatherXM, I contributed to the development of operational systems for evaluating and improving weather forecasts using observational data from more than 8,000 real-time stations. My work included designing and developing benchmarking tools for forecast validation, engineering a hyperlocal forecasting pipeline, and refactoring Python-based quality control mechanisms for meteorological timeseries, resulting in a 50% performance improvement. I also built visualisations to communicate complex weather data to both technical and non-technical stakeholders, and mentored a junior data scientist on scientific software engineering practices. Alongside this, I completed an AI/ML internship at the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT), where I worked with large-scale satellite and radar datasets, applying deep learning methods such as Vision Transformers and CNNs. My work focused on scalable training on HPC systems and the development of reproducible, cloud-native machine learning workflows using containerised pipelines, automated orchestration and experiment tracking. I have also worked as an independent consultant on projects related to HPC network design within the EuroHyPerCon initiative of the EuroHPC JU, as well as on the development of a Climate Risk Assessment Platform.
National and Kapodistrian University of Athens
Athens, Attikí, Greece
Informatics and Telecommunications