About Me
Hello, I'm Samko and welcome to my website. I'm a student at Charles University, where I'm pursuing the bachelor's degree in Computer Science. My main fields of interest are machine learning, mathematics, and theoretical computer science. On this website you can find more information about me and projects I've worked on.
Education
Charles University
Computer Science
2024 -
- Successfully completed the first term and now fighting the second one
Gymnázium Metodova 2
Grammar School
2016 - 2024
- Graduated with straight A's (1.0 GPA)
- Participated in several mathematics and physics competitions
Projects
Matrix Library
A build-from-scratch python library that performs basic matrix operations such as multiplication, REF/RREF form. Apart from that, it can also find the inverse matrix, compute the rank and more.
Data Classification in PyTorch
I worked on on basic classification models to separate both synthetic datasets like circles and blobs from scikit-learn, as well as real-world image classification. I built and trained these models from scratch in PyTorch, focusing on hands-on experience with core machine learning techniques.
NTTS2025 - Using CI/CD methods for continuous benchmarking of statistical methods
I investigated the use of CI/CD pipelines for continuous benchmarking of statistical models. This required preparing a range of models, such as Autoregressions, Moving Average, Exponential Smoothing, and ARIMA, while also developing a framework to automatically run benchmarks with every commit
Outplan
Outplan was built by me and my team at Hack Košice 2024 event. It's a web application that keeps outdoor explorers updated about current weather situation and gives them recommendations on what to bring outside. This project won IBL's prize at Hack Košice 2024.
European Statistics Competition
My team and I had the honor of participating in the European Statistics Competition, where we finished in the top 20 out of over 400 teams. We won the first phase and in the second, we developed a functioning VAR model to capture dynamic relationships between various indicators.
European Statistics Awards - Web Intelligence Challenge
Out of numerous participants, my team and I attained the noteworthy achievement of securing the 1st place in AccuracyPlus Award and 8th pace in Accuracy Awards in the Eurostat's Web Intelligence challenge. The challenge primarily focused on the complex task of identifying and eliminating duplucates amongst job offers.