György Kovács
Business Technology Solutions Implementation & Support Engineer
Budapest, Central Hungary
Summary
👤
Senior🎓
Top SchoolGyörgy Kovács is a seasoned telecommunications and VoIP engineer with 14+ years building and operating carrier-grade IP networks and wholesale voice platforms across Hungary. He combines deep network operations and Level 3 NOC expertise with product and solution architecture experience, having led VoIP core deployments, interoperability testing, and technical-commercial integrations at operators like Vodafone and GTS. Now focused on business technology solutions implementation and support, he bridges technical delivery and customer-facing contract/solution design. György also contributes to open-source ML tooling—adding oversampling methods and classifier integrations to a prominent SMOTE variants repository—reflecting a rare blend of telecom systems know-how with practical software and data-science skills. Based in Budapest, he brings pragmatic problem-solving, regulatory compliance experience, and a habit of building internal tools that streamline operations.
13 years of coding experience
29 years of employment as a software developer
BSCE, Microwave Radio and Space Telecommunications Engineering, BSCE, Microwave Radio and Space Telecommunications Engineering at Kando Kalman Technical CollegeElectrical Technician, Electrical Technician at Kolos Richard Technical School
Hungarian, English
Github Skills (4)
machine-learning10
scikit-learn10
python10
scikit10
Programming languages (4)
Github contributions (5)
analyticalmindsltd/smote_variants
Oct 2018 - Dec 2022
A collection of 85 minority oversampling techniques (SMOTE) for imbalanced learning with multi-class oversampling and model selection features
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:23 releases, 3 reviews, 453 commits in 4 years 2 months
Contributions summary:György appears to have made modifications to the Python code within the `smote_variants` repository. Their changes focus on adding new oversampling methods, and adjusting the parameters for existing ones. The user integrated new classifiers (e.g., GaussianNB, DecisionTreeClassifier) and made changes to existing classes, and refactored some existing functions. These changes indicate a contribution that involved both back-end development and machine learning model implementation.
techniquessmotebinary-classificationimbalanced-datamodel-selection
FalseNegativeLab/mlscorecheck
Oct 2023 - Dec 2024
Testing the consistency of binary classification performance scores reported in papers
Contributions:3 releases, 10 PRs, 154 pushes in 1 year 2 months
accuracyclassificationconsistencyehgf1
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.