Willi Richert
VP Tech at flexa
Greater Munich Metropolitan Area Germany
Summary
🤩
Rockstar🎓
Top SchoolWilli Richert is an experienced engineering manager specializing in mapping, leading teams at Lyft to build on-device and offline maps using computer vision and telemetry. With over a decade in software and more than 15 years of broader engineering leadership experience, he blends hands-on ML and systems expertise with people management to scale reliable mapping pipelines. He has contributed to high-profile open-source work in deep learning—improving core operators and stability in Microsoft's CNTK—bringing a strong mathematical rigor to model implementations. Based in the Greater Munich area, he excels at turning complex sensor and CV data into production-ready mapping solutions and mentoring teams to deliver them. An understated strength is his focus on aligning algorithmic correctness with real-world performance constraints, ensuring models behave predictably in edge cases.
10 years of coding experience
11 years of employment as a software developer
Diplom Informatik, Diplom Informatik at Paderborn UniversitySemester abroad Computer Science, Semester abroad Computer Science at University of Helsinki
German, English
Stackoverflow
2,861reputation
1.4mreached
63answers
13questions
python
top-5%
Github Skills (12)
debugging10
machine-learning10
cntk10
deep-learning10
python9
performance-optimization9
windows6
perl6
forloop6
emacs6
fileio6
combinations6
Programming languages (2)
Github contributions (5)
microsoft/CNTK
Mar 2016 - Sep 2017
Microsoft Cognitive Toolkit (CNTK), an open source deep-learning toolkit
Role in this project:
ML Engineer
ML EngineerContributions:927 commits, 2 PRs, 794 pushes in 1 year 6 months
Contributions summary:Willi contributed to the Microsoft Cognitive Toolkit (CNTK) repository, specifically focusing on improvements related to machine learning and deep learning functionalities. Their commits involve debugging and enhancing the performance and stability of models. The user's work is focused on optimizing and correcting implementations of core operators, and ensuring their accuracy and alignment with the mathematical principles and mathematical logic being used.
cntkdeep-learningcognitive-toolkitmachine-learningdeep-neural-networks
wilrich-msft/keras
Feb 2016 - Mar 2016
Contributions:6 pushes, 1 branch in 5 days
pythondeep-learningrecurrent-neural-networkstheanoneural-networks
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.