Personal site and portfolio

Machine learning, high-performance computing, and practical software engineering.

I'm Petr Kladov, a master's student at VSB - Technical University of Ostrava, specializing in Computational Methods and High-Performance Computing.

My work combines machine learning, numerical methods, parallel programming, and hands-on experimentation on the Karolina computing cluster, with projects spanning research, performance-oriented computing, and software deployed in production-like environments.

LinkedIn GitHub Telegram

Right now I'm focused on neural network research, my master's thesis on alternatives to classical MLPs, and a video automation project built around modern AI tooling.

Overview

Research-minded work with practical implementation

Current focus

My master's thesis compares Multilayer Perceptrons and Kolmogorov-Arnold Networks across predictive performance, memory efficiency, computational cost, training time, and inference requirements.

HPC and numerics

My studies and projects gave me hands-on experience with OpenMP, MPI, CUDA, Slurm-based benchmarking, and experiments on the Karolina computing cluster and related HPC systems.

Applied engineering

I enjoy turning technical ideas into usable systems, from this Django-based personal site and MLflow infrastructure to current automation work around short-form video generation.

Selected work

Projects and work that show range

Master thesis

MLP vs. Kolmogorov-Arnold Networks

Research comparing Multilayer Perceptrons and Kolmogorov-Arnold Networks across predictive quality, memory requirements, computational cost, training time, and inference efficiency.

PyTorch, MLflow, scikit-learn

See LinkedIn profile

HPC project

Parallel Gram-Schmidt process

A university project focused on parallelizing Gram-Schmidt efficiently with OpenMP, MPI + OpenMP, CUDA (+ HIP) followed by experimental tuning and benchmarking.

OpenMP, MPI, CUDA, Slurm

See LinkedIn profile

Web and infrastructure

Personal website and ML tooling

This site is part of a broader personal stack built with Django, PostgreSQL, Docker, NGINX, and Gunicorn, deployed on Ubuntu alongside my MLflow server.

Django, PostgreSQL, Docker, NGINX, Gunicorn

You`re already here!

Background

Links and context

LeetCode Strava Pinkbike