Vladyslav Kostiuk
Data Engineer
Seattle, Washington, United States
Summary
👤
Senior🎓
Top SchoolVladyslav Kostiuk is a Data Engineer in Seattle with nine years of experience designing and operating machine learning and big-data pipelines that bridge research models and production systems. He has built robust ETL and data-integration frameworks at scale, standardized customer onboarding for multiple products, and serves as a first responder for production incidents. His background spans hyperspectral imaging, document-level NLP at multi-million-document scale, and computer-vision model implementation—evidenced by a TensorFlow PSPNet semantic-segmentation contribution that improved multiscale prediction robustness. Comfortable across Python and JavaScript, he combines hands-on ML engineering with pragmatic data ops practices to ensure reliability and observability in critical pipelines. Collected experience at startups and in finance gives him a track record of turning experimental models into maintainable, production-ready systems.
9 years of coding experience
3 years of employment as a software developer
Bachelor's Degree, Computer Science, Bachelor's Degree, Computer Science at University of California, BerkeleyAssociate of Science - AS, Computer Science, 3.9, Associate of Science - AS, Computer Science, 3.9 at City College of San Francisco
English, Ukrainian
Github Skills (7)
semantic-segmentation10
keras10
computer-vision10
tensorflow10
python10
image-processing10
machine-learning9
Programming languages (4)
Github contributions (5)
Vladkryvoruchko/PSPNet-Keras-tensorflow
Jun 2017 - Sep 2019
TensorFlow implementation of original paper : https://github.com/hszhao/PSPNet
Role in this project:
ML Engineer
ML EngineerContributions:60 commits, 23 PRs, 58 pushes in 2 years 3 months
Contributions summary:Vladyslav focused on implementing and refining a TensorFlow-based semantic segmentation model (PSPNet) within the repository. Their contributions included adding script files, cleaning up code, and committing caffe-tensorflow modifications to a PSPNet BN layer. Further work involved integrating sliding/multiscale prediction methods and addressing a multiscale bug, indicating a focus on improving the model's performance and usability.
deep-learningpspnettensorflow
KMDPartners/json-logic-py
Dec 2019 - Aug 2020
Build complex rules, serialize them as JSON, and execute them in Python
Contributions:6 pushes in 8 months
pythonserializecomplex-rulesrulesjson
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.