Volodymyr Halian
AI & Machine Learning Developer
Linz, Austria · Focus: Machine Learning Engineering
Profile
Artificial Intelligence student at Johannes Kepler University Linz, focused on machine learning, deep learning, and practical software development. Academic interests include PyTorch, neural network architectures, optimization, and careful model evaluation. Seeking an internship or working-student opportunity.
Education
Johannes Kepler University Linz
Bachelor's Program in Artificial Intelligence · In Progress
Relevant fields: machine learning, deep learning, probability and statistics, linear algebra, optimization, reinforcement learning, algorithms, and software engineering.
Technical toolkit
- Python, PyTorch, NumPy, Pandas, Matplotlib, scikit-learn
- Coursework: FNN, CNN, backpropagation, gradient descent, model evaluation
- Foundations: linear algebra, calculus, probability, statistics, optimization
Academic studies & explorations
FNN vs CNN
Exploring how architectural inductive biases shape learning on a sequence-based latch task.
Inside the training loop
Connecting neurons, gradients, and parameter updates to the mechanics of learning.
Seeing data differently
From high-dimensional features to interpretable projections with principal component analysis.
Finding structure without labels
Investigating hierarchical clustering, distance metrics, and the structure hidden in unlabeled data.
Practical engineering & business projects
Mousse Bakery — digital projects, AI-assisted workflows & operations
Website-related development, product catalog organization, customer communication, business-rule validation, wholesale coordination, and workflow design.
Live business website: moussebakery.com
Mobile calendar application — prototype
React Native / Expo exploration of a weekly calendar and task-entry interface. Voice entry and schedule-photo extraction remain experimental goals.
Research interests
Causal vs Full Attention — upcoming seminar and practical work at JKU Linz.
JEPA and world models — research interest; no completed implementation claimed.