I am Matteo Gätzner, a Machine Learning Engineer and Researcher with an M.Sc. in Statistics from ETH Zürich and a B.Sc. in Computer Science from TU Berlin. My expertise spans the entire machine learning lifecycle, from building efficient data infrastructure and training complex deep learning models to deploying scalable AI services into production.

By combining a strong foundation in computer science with the rigorous statistical methodologies required for probabilistic inference and uncertainty quantification, I focus on making AI systems both highly performant and highly reliable for real-world applications.

I am a strong advocate for open-source software, clean code, and high-performance computing. I author and maintain several projects that reflect my interest in bridging the gap between research and engineering:

• mini-mcmc: A high-performance MCMC sampling library built in Rust. It implements a suite of advanced algorithms, including HMC, NUTS, and Gibbs sampling, designed for robust probabilistic modeling.
• epub2anki: An open-source library that automatically generates high-quality Anki flashcards from EPUB documents using LLMs, providing a highly cost-effective study tool.

Professionally, I have engineered end-to-end machine learning solutions across multiple domains. My experience includes developing deep-learning models and deploying them as Docker-containerized REST services, managing spatial data pipelines with PostgreSQL, and building optimized ML predictors while automating complex laboratory workflows.

For a comprehensive overview of my professional background, publications, and technical skills, please see my full resume.

I'm always open to discussing engineering challenges, deep-tech projects, or potential collaborations. Feel free to reach out via email at matteo.gatzner@gmail.com or connect with me on LinkedIn and GitHub.