Varis ShieldHeart

I’m a Mathematics student with four years of Python programming experience, focused on building machine learning and software engineering projects. I develop end-to-end ML pipelines—from data preprocessing and feature engineering to model selection, hyperparameter tuning, and evaluation using Scikit-learn and TensorFlow. I care about writing clean, modular, and maintainable code while applying machine learning to real-world problems.

In my projects, I built a dropout-risk predictor with an end-to-end workflow that included feature scaling, careful train/validation/test splitting, and RandomOverSampler while preventing data leakage. I tuned models (KNN, Ridge Classifier, SVC) using GridSearchCV and trained a neural network in TensorFlow with Adam optimization, Binary Cross Entropy, and Early Stopping—achieving 91.6% accuracy and 89.0% F1-score. I also created a Crypto Tracker portfolio manager using the Binance REST API with persistent JSON storage, input validation, exception handling, and modular project structure for maintainability.