Abdullah Makkeh
Guest Researcher at Max Planck Institute for Dynamics and Selforganization
Göttingen, Lower Saxony, Germany
Summary
👤
Senior🎓
Top SchoolAbdullah Makkeh is a computer scientist and mathematician with a decade of experience translating information theory and high-performance computing into practical, interpretable AI systems. As a Guest Researcher and former Senior Scientist in Göttingen, he has led interdisciplinary teams to improve pattern-recognition models, develop autonomous learning paradigms, and publish in top venues such as PNAS and NeurIPS. His technical toolkit centers on Python and PyTorch, and he has advanced novel metrics for disentanglement, stopping criteria for LLMs, and autonomy evaluation in reinforcement settings. Abdullah’s work bridges theory and open-source practice—he has developed analysis tools for time series and neural data and fostered collaborations with industry partners like Honda Research Institute Europe. Fluent in English and German, he combines project leadership with hands-on research, having supervised students from BSc to PhD level. An underappreciated strength is his focus on measurable, performance-driven gains (e.g., a documented 10% boost in pattern recognition models) that make research directly actionable.
10 years of coding experience
7 years of employment as a software developer
MSc Mathematics, MSc Mathematics at Lebanese University - Faculty of SciencesPhD Mathematics and Computer Science, PhD Mathematics and Computer Science at University of Tartu
English, Arabic, German
Github Skills (23)
information-theory10
state-space9
x-plane9
overlapping8
theory8
dynamical-systems8
dynamics8
mutual-information8
ecos6
pid6
dm6
union6
sat5
inference5
time-series5
Programming languages (3)
Github contributions (5)
Abzinger/BROJA_2PID
Oct 2017 - Dec 2019
Contributions:1 release, 71 commits, 63 pushes in 2 years 2 months
Abzinger/SxPID
Mar 2020 - Jan 2024
A differentiable measure of shared mutual information via overlapping exclusions in event (measure) spaces for discrete and continuous variables
Contributions:5 PRs, 31 pushes, 2 branches in 3 years 11 months
differentiablecontinuousdiscretespacesmeasure
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.