Tim Nonner
Principal Applied Scientist at Zalando
Zurich, Zurich, Switzerland
Summary
👤
Senior🎓
Top SchoolTim Nonner is a Principal Applied Scientist based in Zurich with 11 years of experience building production-grade ML pipelines, focusing on NLP, recommender systems, and fraud detection. With a PhD in Mathematics and Computer Science and a background in optimization from IBM Research, he combines rigorous algorithmic thinking with pragmatic product impact—aligning training data to business metrics to halve error rates in key systems. He has led central data science teams, mentored peers, and lectured on algorithmic decision making, bringing academic rigor into industry practice. An active contributor to the popular flair NLP project, he optimized embeddings for memory efficiency by integrating a SQLite-backed WordEmbeddingsStore, demonstrating attention to engineering trade-offs at scale. Tim is a metrics-driven practitioner who routinely verifies A/B-tests and prioritizes Pareto gains that move products and bottom lines.
11 years of coding experience
18 years of employment as a software developer
PhD, Mathematics and Computer Science, PhD, Mathematics and Computer Science at The University of FreiburgDiploma/Master, Mathematics, Diploma/Master, Mathematics at Universität StuttgartTerm abroad, Term abroad at Uppsala universitet
German, English, French
Github Skills (9)
pytorch10
word-embeddings10
machine-learning10
sqlite10
python10
natural-language-processing10
database9
sequence-labeling9
databases9
Programming languages (10)
Github contributions (5)
flairNLP/flair
Dec 2019 - Sep 2020
A very simple framework for state-of-the-art Natural Language Processing (NLP)
Role in this project:
Back-end Developer
Back-end DeveloperContributions:6 commits, 4 PRs, 6 comments in 9 months
Contributions summary:Tim focused on optimizing the `flairnlp/flair` repository for memory efficiency by replacing word embeddings with a database-backed solution using SQLite. Their contributions included the creation of a `WordEmbeddingsStore` class to manage the database and implement methods to load and embed word vectors. The user also integrated the `WordEmbeddingsStore` within the existing `flair` framework and addressed various integration issues. Further improvements included adding a `get_names` method and cleanup of the code.
nlppytorchnamed-entity-recognitionartsemantic-role-labeling
timnon/pyschedule-notebooks
Aug 2015 - Mar 2018
Contributions:13 commits, 11 pushes, 1 branch in 2 years 6 months
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.