Xhensilda Allka

Research Engineer at the Barcelona Supercomputing Center (BSC), one of Europe's leading HPC and AI research institutions. Currently developing a multigrid solver using CUDA GPU programming, combining high-performance computing with machine learning optimization techniques.

Working at the intersection of numerical computing, GPU acceleration, and applied machine learning — delivering performance improvements on large-scale scientific computing problems.

I hold a PhD (Cum Laude) in Data Science from the Universitat Politècnica de Catalunya. My doctoral research focused on applying machine learning and deep learning models to improve the quality of data collected by low-cost IoT sensors used in air quality monitoring systems.This work involved developing and implementing state-of-the-art models — including LSTM, BiLSTM, Transformers, and Gaussian Processes — for sensor calibration, anomaly detection, and outlier detection in real-world environmental sensor networks. The research was published in several IEEE journals and contributed to EU-funded projects in IoT and environmental monitoring.

Assistant Professor at two universities in Albania, teaching technical and mathematical subjects to engineering students. Courses delivered include Numerical Methods for Engineers (Matlab/Octave), Probability and Statistics, and Graph Theory.

This role strengthened my ability to communicate complex technical concepts clearly and accessibly — a skill I bring directly to my work with clients and research teams. Teaching these subjects at university level required deep mastery of the underlying mathematics and the ability to make it practical and applicable for students.