Євген

Evgeny Nevodnich
Data Scientist | Machine Learning/AI Engineer | Analyst

Location: Ukraine | Email: [open contact info](look above in the "contact info" section) | Phone: [open contact info](look above in the "contact info" section)

Additional resources:
kaggle.com/evgenynevodnich | github.com/jenyafross | [open contact info](look above in the "contact info" section)

About
Adaptable IT professional with over 10 years of experience in system
administration, technical support, and programming. Since 2021 I have
been expanding this technical foundation through studies in Math, Data
Science, and Machine Learning.

Technical Skills:
Languages:
Proficiency: Python
Familiar: SQL, JavaScript, HTML/CSS, C/C++, PHP
Tech Stack:
PyTorch, MLflow, Scikit-Learn, Pandas, NumPy, Matplotlib, Jupyter,
LightGBM, CatBoost, XGBoost, Cursor

Experience

Studying, Self-Education Apr 2021 - Present
Advanced Training: Deepening foundations in mathematics and machine
learning through self-study of research papers, academic literature, and
specialized courses.
Competitive Data Science: Secured 3rd place in the Kaggle Playground
Series (S4E12) and achieved 10th and 11th places in Ukrainian Kaggle
competitions (2024).
Architecture Implementation: Implemented deep learning architectures
from scratch, including GPT-2, Vision Transformers (ViT), and DINO.
Current Focus: Actively exploring Large Language Models (LLMs),
Retrieval-Augmented Generation (RAG), and Multi-Agent systems to solve
complex tasks.

Key ML Projects & Applied Domains
Computer Vision:
Implemented image segmentation models using U-Net architecture and
built image classification pipelines.
Developed object counting systems utilizing YOLO and OpenCV.

Audio & Speech Processing:
Developed sound classification models for insect recognition.
Built an end-to-end Text-to-Speech (TTS) pipeline for an English tutor
application.

Information Retrieval & Search Systems:
Quote Search Pipeline: Collected, processed, and indexed movie subtitles in
Elasticsearch to enable precise quote lookup.
Semantic Movie Search: Built a movie search pipeline. Utilized text
summarization and Late Chunking to generate context-enriched
embeddings from raw subtitles, storing and querying them via
PostgreSQL(pgvector).

Time-Series & Tabular ML:
Built predictive financial models using time-series data from crypto stock
markets.
Solved tabular classification and regression tasks using GBDTs (LightGBM,
CatBoost, XGBoost) and Scikit-Learn.

Freelance Sep 2022 - Feb 2026
System administration tasks.

MegaHoster.Network Apr 2013 - Feb 2022
Tech Stack
DevOps: NGINX, Apache, PM2, Proxmox (KVM), OpenVPN, Firewall
Configuration, DirectAdmin, ISPmanager, VestaCP.
Backend & Databases: Node.js, Python, PHP | PostgreSQL, MySQL,
MariaDB, MongoDB.
Frontend: JavaScript, React, React Native, Redux, Next.js, HTML5, CSS3,
Bootstrap.

Key Experience & Achievements
Server Administration & Virtualization: Administered 200+ servers, a
portion of which were used for KVM virtualization hosting over 100 client
environments.
Workflow Automation: Implemented automation scripts using Selenium,
saving up to 8 hours of manual tasks per week.
Full-Stack Development: Developed and deployed whois service from
scratch with React.js, Node.js, and MongoDB.
Legacy Refactoring: Rewrote the frontend megahoster.net website using
Bootstrap, ensuring full cross-device responsiveness and improved UI/UX.
API Integration: Created PHP plugins for WHMCS that integrated with
datacenter APIs, allowing users to check server status, reboot, and reinstall
operating systems.
Infrastructure Optimization: Configured, supported, and optimized
high-availability web services based on LAMP/LNMP stacks and managed
server virtualization via Proxmox.
Feature Development: Developed custom features and bug fixes using
PHP, JavaScript, and CSS based on client requests.