Panagiotis Giannoulis

Led ML development at a healthcare technology company. Designed and deployed machine learning solutions that estimate vital signs (heart rate, respiratory rate, motion, human presence, and sleep stages) using radio frequency (RF) signals from contactless sensors.

Processed user behavior data and applied XGBoost for classification and pattern detection. Built fraud detection models to uncover coordinated abuse of promotional offers by multi-account users.

Developed a Retrieval-Augmented Generation (RAG) pipeline for a clinical assistant chatbot for pediatricians. This system enhanced the chatbot's ability to provide accurate and relevant information.

Built a phone recognition system using multi-angle images and computer vision to identify smartphone models. This enabled automatic estimation of device resale value within mobile exchange kiosks.

Conducted research as a Research Assistant on various European Research Projects including DIRHA and BABY-ROBOT. Contributed to advancements in relevant fields through collaborative research efforts.

Served as a Teaching Assistant for the "Digital Signal Processing (DSP)" course at the Electrical and Computer Engineering department. Provided support to students and assisted with course material.

Completed a PhD in Electrical and Computer Engineering, focusing on multi-sensor processing of the acoustic environment. The thesis was advised by P. Maragos and G. Potamianos.

Earned a Bachelor of Science in Electrical and Computer Engineering. The bachelor thesis focused on speech emotion recognition, advised by P. Maragos and G. Potamianos. Received an award for the highest score in University Entrance Exams.