Summary
Paweł Korus is an applied scientist specializing in computer vision, multimedia signal processing and security, with over a decade of experience bridging academic research and industry productization. Currently at Amazon, he focuses on media integrity across the acquisition and distribution pipeline—camera ISP, learned codecs, forensics and detection of manipulated or synthetic content. His background spans PhD-level work in telecommunications, faculty roles, and postdoc research that produced open-source toolboxes (notably neural-imaging and multiscale-prnu) used for differentiable camera/codec simulation and sensor-fingerprint forensics. He has published in top venues and built practical systems—from Android camera prototypes to exam-grading and web reporting tools—demonstrating an unusual blend of theoretical modeling and hands-on engineering. Paweł also explores adversarial and security facets of ML (dictionary attacks on biometrics, adversarial sensors), combining domain expertise in electronics with a preferred Python + TensorFlow stack.
10 years of coding experience
16 years of employment as a software developer
Ph.D. Telecommunications, Ph.D. Telecommunications at AGH University of Krakow
Polish, English, German, Chinese