Konstantin Chernyshev
Research Scientist at Toloka
Berlin, Germany
Summary
👤
Senior🎓
Top SchoolKonstantin Chernyshev is a research scientist based in Berlin with 8 years of experience focused on LLM evaluation, agentic pipelines, model compression, XAI and pruning, currently driving R&D at Toloka. He blends strong academic training in computational linguistics and machine learning with hands-on engineering—evidenced by maintenance contributions to NLTK’s BLEU implementation to ensure Python 3.12 compatibility. Konstantin also leads Python instruction at Yandex School of Data Analysis, building courseware and automated solution checkers that scale technical education. Comfortable moving between research and production, he has a track record of shipping robust NLP tooling and optimizing models for efficiency while staying engaged in open-source and teaching.
8 years of coding experience
2 years of employment as a software developer
Honours Bachelor's degree, Computer Science, GPA 9.1, Honours Bachelor's degree, Computer Science, GPA 9.1 at Higher School of EconomicsErasmus Mundus European Masters Program in Language and Communication Technologies (LCT), Computational Linguistics, Erasmus Mundus European Masters Program in Language and Communication Technologies (LCT), Computational Linguistics at University of GroningenErasmus Mundus European Masters Program in Language and Communication Technologies (LCT), Computational Linguistics, Erasmus Mundus European Masters Program in Language and Communication Technologies (LCT), Computational Linguistics at Universität des SaarlandesMaster's degree, Mathematic, Computer Science, Machine Learning, Master's degree, Mathematic, Computer Science, Machine Learning at Yandex School of Data Analysis
Github Skills (5)
nltk10
nlp10
python10
natural-language-processing10
machine-learning9
Programming languages (10)
Github contributions (5)
nltk/nltk
Dec 2023 - Dec 2023
NLTK Source
Role in this project:
Back-end Developer
Back-end DeveloperContributions:1 comment in 1 day
Contributions summary:Konstantin primarily focused on improving the `nltk/translate/bleu_score.py` module within the NLTK library. Their contributions included refactoring the `Fraction` class for compatibility with Python 3.12, fixing style issues, and enhancing the code. They also updated the `setup.py` file to reflect the support for Python 3.12. These changes indicate a focus on code maintenance and ensuring the library's compatibility.
nlppythonmachine-learningnltknatural-language-processing
k4black/codebleu
Jun 2023 - Jan 2025
Pip compatible CodeBLEU metric implementation available for linux/macos/win
Contributions:15 releases, 3 reviews, 56 PRs in 1 year 7 months
code-evaluationcode-generationcodebleuevaluationevaluation-metrics
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.