Taras Savchyn
PhD Candidate
Berlin, Germany
Summary
🤩
Rockstar🎓
Top SchoolTaras Savchyn is a PhD candidate and bioinformatics-oriented software engineer based in Berlin with eight years of experience bridging computational biology and machine learning engineering. Trained in cellular biology (BSc/MSc from Taras Shevchenko National University of Kyiv), he now applies that domain expertise at MDC-BIMSB while contributing code and DevOps improvements to notable open-source ML tooling such as PyTorch Ignite. His work on ignite.handlers added typing, robust exception handling, tests for ModelCheckpoint, and GitHub Actions-backed Docker publishing—demonstrating a rare blend of research insight, production-ready engineering, and CI/CD fluency. Colleagues value his ability to translate experimental biology problems into maintainable ML pipelines and reproducible deployments.
7 years of coding experience
Master's degree, Cell/Cellular Biology and Histology, Master's degree, Cell/Cellular Biology and Histology at Taras Shevchenko National University of Kyiv
Ukrainian, Polish, English
Stackoverflow
8,543reputation
1.1mreached
136answers
2questions
pytorch
top-1%
python
top-1%
deep-learning
top-1%
machine-learning
top-1%
Github Skills (24)
pytorch10
github-ci10
python10
machine-learning10
deeplearning-ai10
deep-learning10
githubaction-workflow10
dockers9
docker9
neural-network9
type-checking8
documentation8
typehinting8
object-detection6
pip6
Programming languages (9)
Github contributions (5)
pytorch/ignite
Oct 2020 - Jan 2022
High-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently.
Role in this project:
ML Engineer & DevOps Engineer
ML Engineer & DevOps EngineerContributions:1 release, 83 reviews, 56 commits in 1 year 3 months
Contributions summary:Taras primarily focused on improving the codebase's maintainability and functionality, specifically within the `ignite.handlers` module. They addressed typing issues, added exception handling, and incorporated tests for the `ModelCheckpoint` handler. Furthermore, the user contributed to setting up and configuring GitHub Actions for building and publishing Docker images for the project, demonstrating DevOps skills. They also updated documentation.
pytorchpythondeep-learninghigh-levelneural-networks
trsvchn/ignite
Oct 2020 - Apr 2023
High-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently.
Contributions:50 PRs, 339 pushes, 97 branches in 2 years 7 months
pytorchdeep-learninghigh-levelneural-networksmachine-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.