Matti Lyra
Senior Machine Learning Engineer at Zalando
Berlin Metropolitan Area Germany
Summary
👤
Senior🎓
Top SchoolMatti Lyra is a Senior Machine Learning Engineer based in Berlin with 14 years of experience building production-grade NLP and ML systems and leading cross-functional teams. He holds a PhD in Computational Linguistics and has driven end-to-end product development at Zalando, scaling internal NLP tooling and forming a new applied science organization that merged work from multiple teams. Hands-on across cloud, Docker, PyTorch and distributed data stacks, he has shipped recommendation and automated booking systems and improved model pipelines in production. An active open-source contributor, he has fixed and extended core functionality in widely used libraries like gensim and scikit-learn, including persistence and sparse-matrix handling improvements. Colleagues describe him as a pragmatic research engineer who turns state-of-the-art models into reliable, iteratively shipped products.
14 years of coding experience
15 years of employment as a software developer
Doctor of Philosophy - PhD Computational Linguistics, Doctor of Philosophy - PhD Computational Linguistics at University of SussexFurther vocational studies Game design & development, Further vocational studies Game design & development at Adulta
German, Finnish, English, Swedish
Stackoverflow
12,978reputation
2.1mreached
307answers
29questions
tensorflow
top-5%
nltk
top-5%
machine-learning
top-5%
python
top-1%
keras
top-5%
pytest
top-5%
Github Skills (30)
gensim10
python10
scikit10
machine-learning10
topic-models10
natural-language-processing10
scikit-learn10
topic-modeling10
matrix9
pytest9
data-mining9
nltk9
tensorflow9
deep-learning9
keras9
Programming languages (7)
Github contributions (5)
piskvorky/gensim
Apr 2015 - Jul 2018
Topic Modelling for Humans
Role in this project:
Back-end Developer
Back-end DeveloperContributions:34 commits, 7 PRs, 27 comments in 3 years 3 months
Contributions summary:Matti primarily contributed to improving the `gensim` library, which focuses on topic modeling and natural language processing. Their work involved modifying the `LdaModel` class, specifically related to saving and loading model states, including adding support for ignoring specific parameters during persistence. They also added tests to ensure the functionality of these changes, fixed Python 2 to 3 compatibility issues and provided documentation. Furthermore, the user added support for slicing a `TransformedCorpus`
pythonword-similarityword-embeddingsdata-miningfor-humans
scikit-learn/scikit-learn
Nov 2012 - Sep 2015
scikit-learn: machine learning in Python
Role in this project:Data Scientist
Data ScientistContributions:5 commits, 1 PR, 9 comments in 2 years 10 months
Contributions summary:Matti contributed to the scikit-learn library by addressing several issues related to machine learning algorithms. They fixed a bug in CountVectorizer, improved the Discrete AdaBoostClassifier by adding an early fail condition, and enhanced the BaggingClassifier to handle sparse matrices for various prediction functions. Furthermore, the user modified the predict_log_proba function to accept sparse matrices, demonstrating expertise in optimizing machine learning functionalities. These contributions likely helped improve the accuracy, stability, and usability of the library.
data-analysispythonstatisticsdata-sciencelearn-machine-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.