Ілля

EDUCATION
2019 - 2023 Computer science (Bachelor)
Kharkiv National University of Radioelectronics

2023 - 2025 Data science (Master's degree)
Kharkiv National University of Radioelectronics

SKILLS
Frameworks Technical Proficiency
PyTorch Deep Learning
Hugging Face Computer
Transformers Vision
Numpy NLP
Pandas Data

ILLIA
Scikit-learn Preprocessing
Matplotlib Data Mining
TensorFlow Machine Learning
CHERNIAVSKYI Keras
JUNIOR DATA SCIENTIST Programming Language Skill group
Python SQL
PROFILE C++ OOP
Math
I am a self-motivated student,
with a desire to learn something
new in deep learning. I have PROJECTS
university experience in writing 2025 One-shot learning (military equipment)
machine learning projects and This project implements One-Shot Learning for
various programming tasks. I recognition military equipment. It leverages
aspire to tackle challenging tasks ResNet18 with Spatial Group-wise Enhancements
and cultivate valuable team (SGE) for feature extraction and employs metric
experience learning strategies such as Triplet Loss. The training
process includes both classification-based pretraining
and metric-based fine-tuning.
LANGUAGES https://github.com/da4nik08/One-shot-learning
Ukrainian - Native 2024 GENERATION OF INSTAGRAM POSTS
Developed a model to generate Instagram posts
English - B1 Intermediate based on specific instructions, emulating the style
of advertising pages. Fine-tuned the pretrained
LLaMA3 model using Hugging Face and LoRa PEFT.
Generated learning prompts with LLaMA and built
a user interface for the agent.
https://github.com/da4nik08/Fine-tuning

2024 SHIP SEGMENTATION
The goal of this project is to locate ships on
CONTACT satellite images. The first step was to process the
 Poltava
image: create a Boolean mask, divide it into smaller
parts, etc. As I was solving the segmentation
 [відкрити контакти](див. вище в блоці «контактна інформація»)
problem, I created and trained a U-Net model. To
 [відкрити контакти](див. вище в блоці «контактна інформація»)
display the metrics graphs, I used Tensorboard.
 [відкрити контакти](див. вище в блоці «контактна інформація») https://github.com/da4nik08/Segmentation
чернявський-04b407260/
 https://github.com/da4nik08
.
COURSE
07.2022 NATURAL LANGUAGE PROCESSING WITH CLASSIFICATION AND VECTOR SPACES
DeepLearning.AI(coursera)
 https://coursera.org/verify/B7BFMGRDXTLW

11.2022 NEURAL NETWORKS AND DEEP LEARNING
DeepLearning.AI(coursera)
 https://coursera.org/verify/C5FXGUH3592F

02.2023 IMPROVING DEEP NEURAL NETWORKS: HYPERPARAMETER TUNING,
REGULARIZATION AND OPTIMIZATION
DeepLearning.AI(coursera)
 https://coursera.org/verify/M9HMDN7EVH7R

05.2023 SEQUENCE MODELS
DeepLearning.AI(coursera)
 https://coursera.org/verify/E2YCBTJGEWHA

07.2023 CONVOLUTIONAL NEURAL NETWORKS
DeepLearning.AI(coursera)
 https://coursera.org/verify/GMZCKMB4YK9P

09.2023 STRUCTURING MACHINE LEARNING PROJECTS
DeepLearning.AI(coursera)
 https://coursera.org/verify/54SCJESZTHZ2

10.2023 GOOGLE CLOUD BIG DATA AND MACHINE LEARNING FUNDAMENTALS Google
Cloud(coursera)
 https://coursera.org/verify/V7UCS778Z64C

11.2023 BUILD BASIC GENERATIVE ADVERSARIAL NETWORKS (GANS)
DeepLearning.AI(coursera)
 https://coursera.org/verify/MVRDG8V37L4U

INTERESTS
Artificial Intelligence
Movies
Computer games
Cycling