Богдан

Bohdan Uhryn
Contact
 Lviv, Ukraine
 [open contact info](look above in the "contact info" section)
 GitHub: github.com/UgrynBohdan
 LinkedIn: [open contact info](look above in the "contact info" section)

Professional Summary
Detail-oriented Computer Science student specializing in Computational Intelligence at Lviv
Polytechnic National University. Proficient in multiple programming languages and technologies with
a strong focus on AI, machine learning, and backend systems. Successfully designed and deployed a
microservice-based AI prediction system using modern tools like Docker, Flask, PyTorch, and
MySQL. Passionate about building smart systems with scalable architecture and intelligent behavior.

Technical Skills
 Languages: C, C++, Java, Python, JavaScript, HTML, CSS
 Frameworks/Tools: Flask, Django, Jupyter, PyTorch, Node.js, Spring, React, Vite, Redis, Git,
Docker, MySQL
 Concepts: Machine Learning, Deep Learning, Data Analysis, Microservice Architecture, REST
APIs, Neural Networks
 Database & DevOps: MySQL, Docker Compose, Redis, GitHub Actions

Projects
GamePredictor – Microservices AI Sports Forecasting Platform
A full-stack AI-powered application for predicting outcomes of football matches using deep learning.
 Built and trained neural network models using PyTorch (accuracy ~78% on validation football
datasets).
 Developed RESTful API with Flask for processing predictions and user data.
 Created authentication microservice (auth_service) with secure password hashing and JWT.
 Integrated data analysis notebooks (.ipynb) for model development and evaluation.
 Frontend built with React + Vite; includes pages for login, registration, and predictions.
 Used Redis for caching model predictions, improving response time by ~40%.
 Fully containerized using Docker and orchestrated with Docker Compose (available on Docker
Hub).
 Designed scalable database schema for storing user data in MySQL.
Estimated System Performance:
 Handles up to 100 concurrent users with <100ms response time via Redis cache.
 Model trained on ~75,000 match records and tested on live football data.
 Deployment size: ~11GB containerized image, auto-scalable with Docker Compose.

Education
Lviv Polytechnic National University
Bachelor in Computer Science – Computational Intelligence of Smart Systems
Expected Graduation: 2026

Languages
Ukrainian (native), English (intermediate), Polish (intermediate)