Applying Machine Learning and Signal Processing Methods in Instrument Classification. Proposed sophisticated augmentation methods for Polyphonic Instrument Recognition, while applied different CNN-LSTM architectures for Instrument Classification through raw waves.Completed my Master's Thesis on "3D Body, Face and Hands Reconstruction with applications on Sign Language Recognition" supervised by Professor Petros Maragos, diving into the field of Computer Vision and gaining expertise not only in state-of-the-art 3D body reconstruction methods, convolutional and recurrent networks, supervised and semi-supervised learning but as well in handling large datasets and big data, combining multiple streams of information, re-implementing academic papers an acquiring writing and presentation skills.

Worked on creating a real-time robotic application for following, identifying, and grabbing grape clusters in a vineyard for the HexaTerra Legged Robot.Experimented with state-of-the-art methods for object detection, i.e. Mask-RCNN and Yolo-v3, while exploiting Cycle-GANS for applying de-raining image recovery to make the system robust.Our work "Robust Visual Servoing for Precision Agriculture Tasks using Learning and De-rain Image Recovery via GANs" has been submitted at ICRA 2022.

Worked as a part-time job for over 5 years teaching Mathematics, Physics and Programming for Junior and High School. Gained extensive teaching experience through the years by teaching 5 very diverse students, one of whom was dyslectic, while increasing the performance of all of them.Prepared a high school student for the nationwide university entrance examination in Mathematics, achieving his highest grade in this subject, and entering the Department of Mathematics at National and Kapodistrian University of Athens.