Suzen Fylke
Deep Learning Engineer at Embedl
Gothenburg, Västra Götaland County, Sweden
Summary
👤
Senior🎓
Top SchoolSuzen Fylke is a Deep Learning Engineer with 10 years of experience building scalable ML pipelines and production tooling, now bringing efficient neural models to embedded systems at Embedl. She spent several years at Twitter designing and operating TFX- and Kubeflow-based pipelines, driving cost-saving GPU migrations and authoring RFCs that influenced cross-team roadmaps. An active open-source contributor and maintainer in the TFX ecosystem and Model Card Toolkit, she improves developer UX by fixing docs and extending custom executor support in widely used projects. Based in Gothenburg and curious about creative coding, Suzen pairs rigorous engineering with a mission-driven focus on helping socially-minded organizations adopt practical AI.
10 years of coding experience
4 years of employment as a software developer
Postbaccalaureate Studies | Bachelor's Degree Computer Science, Postbaccalaureate Studies | Bachelor's Degree Computer Science at The City University of New YorkBachelor's Degree American Studies, Bachelor's Degree American Studies at Columbia University
nyanja, Latin, Swedish, Italian, English
Github Skills (5)
machine-learning10
tensorflow10
python10
apache-beam9
documentation8
Programming languages (10)
Github contributions (5)
tensorflow/tfx
Jul 2020 - Oct 2022
TFX is an end-to-end platform for deploying production ML pipelines
Role in this project:
ML Engineer
ML EngineerContributions:11 commits, 4 PRs, 27 comments in 2 years 3 months
Contributions summary:Suzen primarily focused on modifying the TFX framework's documentation and codebase related to custom executors and graph regularization. They updated docstrings to reflect changes in class specifications and corrected example usages for custom executor specifications in multiple components, including `Trainer`, `example_gen`, and `transform`. Additionally, the user's commits included merging branches related to documentation fixes for custom executor specifications and modifications related to the handling of data views and data formats in the transform executor.
deployingend-to-endml-pipelinesmlmlops
hunterwics/hunterwics.github.io
Apr 2017 - Nov 2021
Contributions:7 pushes, 1 branch in 4 years 7 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.