Presenter

Konstantinos Parasyris is a computer science researcher and a systems software enthusiast. His research empowers HPC application scientists to perform research faster and more easily. His work in compiler optimizations, programming languages, parallel computing, approximate computing, and performance optimizations empowers HPC scientific developers to effortlessly develop efficient scientific applications. His programming model, HPAC, allows developers to describe software approximations in the application source code. His research project, LocalExperts, is developing mechanisms to automatically identify ill-performing code locations. He co-designed the AMS software that allows HPC applications to replace parts of their computations with machine learning models. His work is frequently published in top conferences (PPoPP, SC, CGO, PACT) and journals (TPDS, TACO), and he is a member of the organizing committee of AI4Dev and the TPDS review board. He received his doctoral degree in computer engineering in 2018 from the University of Thessaly, Greece.