Віктор

Makoziuk Viktor
§ moon-1945 | # [відкрити контакти](див. вище в блоці «контактна інформація»)
Student at Lviv Polytechnic National University. Passionate about data science with a strong foundation in
math, programming, machine learning, and analytics. Seeking opportunities to contribute to data-driven
projects and help teams achieve impactful results while growing professionally.

Experience
08.2024-02.2026 ML engineer in Compatibl. I’m part of a development team that implements scientific
papers on financial mathematics into software. Tech stack: Python, NumPy, Torch,
Pandas, Matplotlib, Plotly

Projects
Wine dataset research Python, Scikit-learn, Matplotlib
I did exploratory data analysis of the wine dataset. I applied various baseline models (logistic and linear
regression, decision trees) to predict the quality of wine. Github: wine-dataset-research
Image Sharpener Python, Torch, Torchvision, Matplotlib
Developed an image sharpening model that reconstructs high-frequency details from blurred 256×256
images. Built a complete training and evaluation pipeline in PyTorch, including data preprocessing,
augmentation, and model validation. Github: image sharpener
Skvork LoRA Python, Torch, Torchvision, Diffusers, Matplotlib
Fine-tuned a LoRA model on Stable Diffusion 1.5 to learn a custom fantasy creature concept from the
Tuxemon dataset. Built data preprocessing, training, and inference pipelines for text-to-image generation
using Diffusers and PyTorch. Github: Skvork LoRA
Bird & Animal Sound Classifier Python, Torch, Torchaudio, Transformers, Scikit-learn
Developed a deep learning framework for audio-based classification of bird and mammal vocalizations.
Built pipelines for training custom CNNs (incorporating residual connections and Squeeze-and-Excitation
attention) and fine-tuning pretrained audio encoders like DistilHuBERT and BirdAVES. Implemented
prototypical networks for few-shot learning to evaluate out-of-domain generalization on mammal audio
data. Github: bird-classification

Education
2022-present BSc in Software Engineering, Lviv Polytechnic National University

Activities
2022-2023 Julia’s Dream 1-year remote internship by Massachusetts Institute of Technology. Mentored
mathematical research project about Chord and Jacobi diagrams. Worked on Birman and
Dolgushev Conjectures
2022 Selection round for International Mathematics Olympiad, Linz, Austria.
Ukrainian Preparatory Camp, for participants of the selection round for the International
Mathematics Olympiad
2021 RMM (Romanian Masters of Mathematics) selection in Kyiv, Ukraine
All-Ukrainian Math Olympiad (2nd place)
2020 All-Ukrainian Chemistry Olympiad (3rd place)
All-Ukrainian Math Olympiad (3rd place)
Presentation at National Stage of Junior Academy of Sciences of Ukraine research on Eco-
nomics about linear inequalities in optimization problems.

Publications
Antonovych Dmytro, Makozyuk Viktor, Tysiachnyi Vladyslav and Semenyakin Mykola (2023).
Matching of frames of open Jacobi diagrams and chord diagrams. PDF Preprint

Skills
OS Linux, Windows
Code Python, C, C++, C#, Git, Docker
Databases MySQL / MS SQL, MongoDB
Analysis Statistics, Linear Algebra
Languages Ukrainian(fluent), English(B2)