I am a Computer Science student focused on understanding machine learning beyond surface-level implementation.
My goal is not only to train models, but to deeply understand how they behave, why they fail, and how to systematically improve them.
I am currently building projects that involve neural network implementation, model evaluation, and architecture experimentation.
- Deep Learning fundamentals
- Convolutional Neural Networks (CNNs)
- Model evaluation (Accuracy, Precision, Recall, F1-score)
- Data preprocessing & augmentation strategies
- Regularization techniques and performance tuning
- Understanding backpropagation and gradient descent
Developing and improving a CNN-based multi-class classification model.
Experimenting with:
- Data augmentation strategies
- Architecture refinement
- Regularization techniques
- Confusion matrix analysis
- Macro F1-score evaluation
Implementing neural network components manually to understand:
- Forward propagation
- Backpropagation
- Gradient descent optimization
- Model convergence behavior
Languages
Python • JavaScript • Java • SQL
Machine Learning & Data
TensorFlow • Keras • scikit-learn • NumPy • Pandas • Matplotlib
Backend & Tools
Node.js • Express • PostgreSQL • Git • Linux
I believe strong machine learning engineers are built through experimentation, debugging, and understanding model failures — not just achieving high accuracy scores.
I aim to continuously improve by refining architectures, analyzing results, and questioning every assumption.
LinkedIn: https://linkedin.com/in/arber-zylyftari
Medium: https://medium.com/@arberzylyftari123