Göktuğ Karakaşlı
Senior Software Engineer at idealworks
Munich, Bavaria, Germany
Summary
👤
Senior🎓
Top SchoolGöktuğ Karakaşlı is a Senior Software Engineer based in Munich with 7 years of experience building robotics, computer vision, and deep learning systems across research and product environments. He blends hands-on embedded and FPGA development with modern ML engineering—shipping ROS-based autonomous robot software, PCIe kernel-user interfaces, and neural optimizer research implemented in PyTorch. At the University of Freiburg he built benchmarks and competition environments for learned optimizers and dynamic scheduler evaluation, reflecting a rare mix of reproducible research and production rigor. An active open-source contributor, he improved optimizer implementations and tests in the popular tinygrad project and optimized performance-critical robotics algorithms in PythonRobotics. Known for pragmatic refactors and vectorized performance gains, he moves projects from prototype to maintainable codebases while minimizing per-task tuning through smarter learning-rate and scheduler design.
7 years of coding experience
5 years of employment as a software developer
Master of Science - MS, Computer Science, 3.3/4.0, Master of Science - MS, Computer Science, 3.3/4.0 at The University of FreiburgBachelor's degree, Electronics and Communication Engineering, 3.26/4.0, Bachelor's degree, Electronics and Communication Engineering, 3.26/4.0 at İstanbul Teknik ÜniversitesiBachelor's degree, Electronics and Communication Engineering, 3.95/4.0, Bachelor's degree, Electronics and Communication Engineering, 3.95/4.0 at Kocaeli Üniversitesi
Github Skills (14)
algorithm10
pytorch10
localization10
pathfinding10
robotics10
planning10
optimisation10
optimizers10
python10
optimization10
numpy9
testing9
slam9
machine-learning9
Programming languages (10)
Github contributions (5)
AtsushiSakai/PythonRobotics
Oct 2019 - Dec 2019
Python sample codes and textbook for robotics algorithms.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:23 commits, 11 PRs, 7 comments in 2 months
Contributions summary:Göktuğ primarily focused on optimizing and maintaining existing code related to robotics algorithms within the PythonRobotics repository. Their contributions included vectorizing obstacle cost calculations in the Dynamic Window Approach algorithm, improving performance. They also merged updates and refactored code within the particle filter localization module, and made improvements related to the resampling calculation. Additionally, they added functionality for rectangle robots and corrected issues.
pythonpybulletcvxpyroboticsmapping
tinygrad/tinygrad
Oct 2020 - Jan 2021
You like pytorch? You like micrograd? You love tinygrad! ❤️
Role in this project:ML Engineer
ML EngineerContributions:1 review, 5 commits, 3 PRs in 3 months
Contributions summary:Göktuğ contributed to the development of the `tinygrad` library, focusing on optimization and testing of the Adam optimizer. They added tests for the optimizers (Adam, SGD, RMSprop) and implemented an efficient version of Adam. Furthermore, the user refactored code, changing imports and removing obsolete arguments from functions.
deep-learningpytorchmicrograd
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.