Назар-Олексій

NAZAR-OLEKSII
UMANTSIV
MACHINE LEARNING DEVELOPER

PROFILE
I love creating something useful — especially where technology can help
people. I am interested in IT, science and technology, medicine, design, and
CONTACT chess. In my free time, I work on pet projects, helping to develop digital
tools for medicine. I believe that technology should work for people —
[открыть контакты](см. выше в блоке «контактная информация») simply, clearly, and effectively.

[открыть контакты](см. выше в блоке «контактная информация») EDUCATION

Uzhhorod, Ukraine
Master of Computer Science 2025 - 2027
Uzhhorod National University
@nazareth1027

Nazareth2710
Bachelor of Computer Science 2021 - 2025
Uzhhorod National University
SKILLS
Python EXPERIENCE
Tensorflow, PyTorch Lung Q 2025
Scikit-Learn Graduation project
Numpy, Pandas Designed and developed a Streamlit web application for diagnosing
Matplotlib, Seaborn lung diseases using ResNet-18 and XGBoost models.
Built dual modules for X-ray classification (normal, pneumonia,
Streamlit tuberculosis, COVID-19, medical artifacts) and survey-based cancer
React risk prediction.
Achieved 94% accuracy (ResNet-18) and 83.5% accuracy (XGBoost);
Express
implemented language/theme switchers and secure result storage.
MongoDB
Tuberculosis Diagnostic System 2024
Git
Term project
Docker
Trained a CNN model for detecting tuberculosis on chest X-rays using
Keras, TensorFlow, and OpenCV.
Reached 91.3% accuracy and F1-score 0.89; integrated the model
LANGUAGES into an intuitive medical user interface.

Ukrainian (Native)
Credit Scoring System 2023
English (Intermediate) Term project

Developed ML models (Decision Tree, Logistic Regression, Random
Forest) on a Kaggle dataset (438 557 records).
Improved class balance using SMOTE; achieved 81.4% accuracy
(Decision Tree) and 80.3% (Random Forest).