All experience
Postdoctoral Researcher
University of Zurich (EBPI)
- Fine-tuned six transformer models to classify Multiple Sclerosis patient narratives into 27 categories, leveraging outputs for LSTM-based disease journey prediction.
- Developed a multilingual text analysis pipeline for raw patient interview transcripts integrating sentiment analysis, emotion detection, topic modeling, and automated PII anonymization.
Doctoral Student -> Senior Researcher
University of Murcia
- My work focused on a data-driven evaluation of users’ capabilities in diverse technology-mediated environments that can generate rich data sets through the users’ interaction.
Research Intern
ETH Zurich
- At Future Learning Initiative, I was analyzing the students’ problem-solving strategies in the introduction to data science course based on the screen recordings in order to understand the interplay of cognition and affect in open-ended and complex problem-solving.
- Ongoing effort towards publishing results.
Data Science Intern
women++
- Together with my team, I worked on reconstructing the OpenEdu.ch educational platform for Switzerland.
- Created an ontological definition that describes the content found at OpenEdu.ch that will improve Wikidata learning material organization and reachability.
Google Summer of Code Student
Machine Learning for Science (ML4Sci)
- Used machine learning techniques to identify relationships between planetary mapped datasets in order to provide a deeper understanding of planetary surfaces and to have predictive power for planetary surfaces with incomplete datasets.
- Built interactive tool for analyzing the Moon. Ongoing effort towards publishing results.
Deep Learning Research Intern
University of Sydney
- Developed a model for detecting various marine life objects based on state the art in Convolutional Neural Networks.
- Achieved accuracy of up to 85% in recognition of the most important groups of objects.
Data Scientist Intern
study.eu
- Redesigned data scraping tool for gathering data of study programmes across Europe and integrated the extracted data in the internal search engine. The number of programs covered by the system expanded from 4.000 to 20.000.
- Data analysis of user interest across programs based on multiple patterns, resulting in valuable insights for program discovery and advertised partners.
Research Intern
Taipei Medical University
A systematic review of novel approaches for applying ML to Medicine with a strong focus on COVID-19 spread prevention.
Research Intern
University of Piraeus
- Developed novel time-series approach for link prediction in temporal graphs relying on Support Vector Machines.
- Achieved accuracy of up to 80% in link prediction relevant datasets such as IMDB.