Machine Learning · Deep Learning · Model Development
I’m a machine learning engineer based in Prague, interested in deep learning, probabilistic modeling and understanding how modern models work. My professional background includes production ML and data systems, while my independent work is increasingly focused on model development, experimentation and learning from first principles.
I’m open to selected remote contract work with small US and European ML and AI teams, particularly part-time roles involving technically challenging problems, experimentation and real engineering ownership.
Independent project · Deep learning
Implemented a decoder-only Transformer from scratch in PyTorch to understand the core mechanics of modern language models, including self-attention, training, evaluation and autoregressive generation.
View projectIndependent project · Probabilistic ML
Built a probabilistic model for professional tennis and evaluated it against bookmaker markets, focusing on calibration, Brier score and decision quality rather than classification accuracy alone.
View projectPrivate client · Production project
Built and delivered a production analytics application connecting sales and product catalog data into a unified workflow for exploring commercial performance and reducing manual reporting work.
Read case studyNeural architectures, representation learning and understanding modern models beyond high-level APIs.
Training, evaluation, ablations and controlled experiments aimed at understanding why models work, where they fail and how they can improve.
Uncertainty, calibration and models designed to support reliable decisions rather than optimize a single benchmark metric.
I’m open to selected remote contract work with small ML and AI teams, particularly part-time engagements involving model development, experimentation and technically challenging engineering.