Silva Bashllari

Summary

Data scientist and junior researcher with a strong focus on bridging the social and technical fields. I bring hands-on experience in Python, SQL, Java, and statistical analysis, with a solid background in working with nationally representative datasets and translating quantitative findings into policy-relevant insights. During my graduate studies at Politecnico di Torino, I developed analytical projects involving clustering, text analysis, and applied statistical methods. I have co-authored two national-level research papers based on large-scale surveys in Albania, including analysis of social indicators and population perceptions. My work combines rigorous quantitative analysis with field-informed understanding gained through engagement with grassroots tech communities and civil society, fostering the use of data and AI.

Technical Skills

• Python & its data science libraries (Pandas, Numpy, Scikit-learn, Matplotlib, Seaborn, PyTorch, NetworkX, etc.)
• SQL, MongoDB, Hadoop, Spark | Git and GitHub | Java | Excel | Tableau | R | Canva

Professional Experience

Data Analyst Trainee - Local Public Health Unit, Vlore, Albania (11/2025–01/2026)

Traineeship. Conducted data analysis to understand long-term patterns of epidemiological diseases, their frequency of occurrence, associated symptoms, and their relationships with demographic data, using Python and various data visualization tools.

External Research Associate & Data Scientist - CCIS, Tirana, Albania (08/2025–11/2025)

Project-based. Collaborated on a project to conduct research and data analysis based on surveys on the usage and trust in data from various nonstate actors, such as businesses, universities etc. in Albania using Python and various visualization tools.

External Research Associate & Data Scientist - Network Albania (05/2025–08/2025)

Project-Based. Collaborated on a project basis as an external research associate and data scientist, by conducting when needed data analysis and serving as an advisor for matters related to AI.

Co-Organizer & Program Curator - LDA Europe, Tirana, Albania (03/2025–05/2025)

Project-based. Organized two international interdisciplinary conferences LeapIn with a focus on how AI and global political changes are shifting multiple disciplines and DreamIn 2025 with a focus on salesforce. For LeapIN, I was also the program curator, given the focus on AI.

Grassroots Community Co-Manager - DAIT, Tirana, Albania (06/2024–05/2025)

Part-time. Organized and taught in workshops and events to simplify AI concepts for general audiences. Fostered collaborations with industry experts and academia.

Assistant Tutor - Programming Courses, Tirana Center of Technology, Albania (10/2020–06/2021)

Part-time. Delivered hands-on programming instruction in Java and Python to beginner-level learners.

Education

• MSc in Data Science and Engineering (Grade: 106/110) - Polytechnic of Turin, Italy (09/2021–12/2025)
• Summer School on Responsible AI and Financial Literacy - European Forum Alpbach, Austria (08-09/2023)
• Academy For Social Democracy - Friedrich-Ebert-Stiftung (FES), Albania (10-11/2020)
• BA in Business Informatics (Grade: 10/10) - Mediterranean University of Albania, Albania (09/2018–07/2021)

GitHub Projects

• Clustering Consumer Spending - Applied unsupervised learning techniques, like K-means, to cluster customers by spending behavior, using Principal Component Analysis (PCA) for dimensionality reduction.
• Federated Learning - Implemented advanced strategies including client selection, domain generalization, and model pruning, with a focus on reducing costs and improving generalization across heterogeneous data.
• Regression Task on Particle Tracking - Developed a multi-output regression model to predict particle coordinates (x, y) from RSD sensor signals, leveraging ensemble models and signal processing techniques to improve spatial accuracy.
• Sentiment Classification of Tweets - Built a binary sentiment classifier using natural language processing techniques with tools such as NLTK and scikit-learn achieving a good F1 metric through Multinomial Naïve Bayes.

Languages

Albanian (native) | English (C2 – TOEFL iBT) | Italian (B2 – CELI3) | French (A2)

Publications, Conferences, European Research Memberships & Awards