I’m a penultimate year PhD student in Machine Learning / Brain-Computer Interfaces (BCIs) at the Department of Computing, Imperial College London, under the supervision of Prof. Stefanos Zafeiriou. My main research interests lie in the intersection of Deep Learning and Brain-Computer Interfaces (e.g., Differentialable Signal Processing, Geometric Deep Learning, Causality).
Since 2021, I have been working as a Machine Learning Engineer at Cogitat - [check out our released Demos] - where I have been developing novel deep learning methods for EEG-based Brain Computer Interfaces (BCIs).
During my PhD, I am also an Academic PhD Student Representative for the Department of Computing, Imperial College London, a Microsoft Student Ambassador, Co-Organizer of Imperial Computing Conference (ICC) and Co-Organizer of London Geometry and Machine Learning (LOGML) Summer School.
My past working experience includes in short: Facesoft (Machine Leaarning / Software Engineering Intern), Daedalean AI (Machine Learning Thesis Project), Smart Power Networks (Data Scientist) and The Mouse Team (Founder - Android Developer).
Prior to joining the Department of Computing as a PhD student, I completed my Master of Engineering (MEng - intergrated Bachelors) at the Department of Electrical and Electronic Engineering, Imperial College London. I undertook my Master’s year at ETH Zürich as a visiting student where I had the honor to conduct my Master Thesis at the Data Analytics Lab under the supervision of Prof. Thomas Hofmann.
Before Imperial, during my gap year, I attended the bachelor’s programme at the Department of Electrical and Computer Engineering, National Technical University of Athens and the Engineering Preparation programme at Oxford Royale Academy (University of Oxford - St. Catherine’s College). As a high school student, I was a national champion and a member of the Greek National Karate team.
To get in touch, you can e-mail me at: konstantinos[dot]barmpas16[at]imperial[dot]ac[dot]uk
[This website is under construction and will be frequently updated]