Sven Kreiss
Computer Vision Engineer at Meta
Switzerland
Summary
🤩
Rockstar🎓
Top SchoolSven Kreiss is a computer vision engineer at Meta Reality Labs with 11 years of experience bridging academic research and product-grade ML systems. He previously led geospatial machine learning and early engineering growth at Sidewalk Labs (Alphabet) and developed production ML pipelines at startups, bringing both startup agility and enterprise rigor. Sven’s background in particle physics (PhD, NYU) and a postdoc at EPFL—during which he created the influential OpenPifPaf pose estimator—gives him uncommon depth in statistical methods and vision research. He contributed to high-profile open-source projects like OpenSfM and core improvements in gensim and OpenPifPaf, emphasizing robustness, type safety, and reproducible training pipelines. Based in Switzerland, Sven combines hands-on engineering with mentoring experience from advising students at EPFL and a rare history of time at CERN and on the core team that discovered the Higgs boson.
11 years of coding experience
3 years of employment as a software developer
Master, Physics with Mathematical Physics, Master, Physics with Mathematical Physics at The University of EdinburghPhD, Elementary Particle Physics, PhD, Elementary Particle Physics at New York University
English, German
Stackoverflow
481reputation
131kreached
5answers
1question
codeigniter
top-5%
Github Skills (25)
pytorch10
gensim10
word2vec10
python10
machine-learning10
typehinting10
performance-optimization10
computer-vision10
type-checking10
nlp10
3d-reconstruction10
debug9
debugging9
opencv9
codeigniter9
Programming languages (4)
Github contributions (5)
openpifpaf/openpifpaf
Feb 2019 - Jan 2023
Official implementation of "OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association" in PyTorch.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:63 releases, 47 reviews, 1713 commits in 3 years 11 months
Contributions summary:Sven primarily worked on the core software components of the "openpifpaf/openpifpaf" project. Their commits indicate a focus on debugging, refactoring, and enhancing various modules within the project's core functionality. The code changes include optimizing efficiency and readability within modules like "network/losses.py" and "show/painters.py", showing a focus on machine learning model training and visualization. Further, the user demonstrated expertise by refactoring and refactoring the datasets and processing for training.
pytorchkeypoint-detectionopenpifpafassociationdeep-learning
mapillary/OpenSfM
Nov 2021 - Jun 2022
Open source Structure-from-Motion pipeline
Role in this project:Back-end Developer
Back-end DeveloperContributions:9 commits, 5 PRs in 7 months
Contributions summary:Sven primarily contributed to improving the codebase's type safety and data structure integrity. Their work involved adding and fixing type annotations in Python files, particularly within the `opensfm/synthetic_data` and `annotation_gui_gcp` directories. Furthermore, they refactored the `GroundControlPoint` struct, removing local coordinate dependencies and streamlining related loading functions. This included changes in `opensfm/dataset.py`, `opensfm/io.py`, and other files. These changes likely enhanced code maintainability and data consistency within the project.
structure-from-motionpythonpipelinemotionintrinsics
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.