Mario Šaško
Software Engineer at Hugging Face
Varaždin, Varaždin County, Croatia
Summary
🤩
Rockstar🎓
Top SchoolMario Šaško is a software engineer with seven years of experience specializing in machine learning engineering and open-source collaboration, currently contributing at Hugging Face from Varaždin, Croatia. He has hands-on expertise with PyTorch and model internals, evidenced by bug fixes and improvements to flagship repositories like transformers, datasets, and evaluate. Mario pairs code contributions with technical writing, improving documentation clarity across major ML tooling—an indicator of both engineering precision and user-focused thinking. His background includes research and teaching roles at FER and industry internships, reflecting a strong academic foundation in computer science and practical experience bridging research and production. Notably, his contributions include migration to modern PyTorch patterns and cross-platform improvements, showing attention to maintainability and reproducible ML workflows.
7 years of coding experience
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at FER
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (16)
markdown10
transformer10
machine-learning10
pytorch10
nlp10
deep-learning10
markdown-it10
python10
evaluation10
documentation10
datasets10
technical-writing10
modeling9
model-driven9
model-driven-development9
Programming languages (8)
Github contributions (5)
huggingface/datasets
Sep 2020 - Jan 2023
🤗 The largest hub of ready-to-use datasets for ML models with fast, easy-to-use and efficient data manipulation tools
Role in this project:
Technical Writer
Technical WriterContributions:603 reviews, 313 commits, 559 PRs in 2 years 4 months
Contributions summary:Mario primarily contributed to improving the project's documentation by addressing typos, fixing formatting issues, and updating links. Their work focused on refining the documentation for various features, including the loading of datasets, format settings, and providing instructions on how to share datasets. The edits show a concern for clarity and accuracy within the existing documentation structure, with contributions made across multiple documentation files, indicating a consistent effort to enhance the overall readability and usability of the project's resources.
ml-modelstensorflownatural-language-processingmanipulationdata-science
huggingface/evaluate
Feb 2021 - Mar 2022
🤗 Evaluate: A library for easily evaluating machine learning models and datasets.
Role in this project:ML Engineer
ML EngineerContributions:48 commits, 2 PRs, 2 pushes in 1 year 1 month
Contributions summary:Mario primarily contributed to improving the codebase's type safety, style, and consistency by fixing smaller issues, removing unused imports, and updating code. They also added cross-platform support for the datasets-cli. Furthermore, the user fixed various typos and grammatical errors in the documentation and replaced legacy torch.Tensor constructor with torch.tensor and updated the logging and caching systems. The user also fixed several issues related to the metrics and feature systems of the library.
evaluateevaluation-frameworkdata-scienceevaluationmachine-learning
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.