Stefan Pantic
Expert Software Engineer (SWE IV) at TomTom
Belgrade, Central Serbia
Summary
👤
Senior🎓
Top SchoolStefan Pantic is an Expert Software Engineer with nine years of hands-on experience building cross-platform SDKs and distributed systems, currently working on TomTom's online Routing API and Navigation SDK. He has deep C++ expertise complemented by mobile (Kotlin, Swift) and ML work, having contributed reinforcement-learning improvements to the high-profile Ray project. Stefan has repeatedly driven architecture and public API design, shipped production routing and PDF editing products, and mentored junior engineers across multiple TomTom roles. His background spans high-performance actuarial engines, cross-platform speech recognition, and distributed neural training pipelines—showing a rare blend of low-level systems thinking and applied ML. Based in Belgrade, he pairs practical delivery focus with a propensity for improving usability in complex algorithmic codepaths.
9 years of coding experience
6 years of employment as a software developer
Bachelor's degree Computer Science, Bachelor's degree Computer Science at University of Belgrade, Faculty of Mathematics
English, Serbian
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (11)
pytorch10
machine-learning10
ray10
deeplearning-ai10
deep-learning10
python10
reinforcement-learning10
tensorflow9
distribute9
data-science8
llm7
Programming languages (4)
Github contributions (5)
ray-project/ray
Feb 2019 - Jul 2019
Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
Role in this project:
ML Engineer
ML EngineerContributions:5 commits, 9 PRs, 15 comments in 4 months
Contributions summary:Stefan contributed to the development and enhancement of reinforcement learning models and related components within the Ray project. They worked on adding custom LSTM detection, integrating action space functionalities within model architectures, and addressing issues related to multi-discrete action spaces. These changes primarily involve modifications to policy graphs and model implementations, showcasing a focus on improving the usability and functionality of RL algorithms within the Ray framework. The user also worked on entropy coefficient decay for improved training.
pythonconsistsruntimetensorflowserving
stefanpantic/dotfiles
May 2017 - Aug 2022
Contributions:54 pushes, 1 branch in 5 years 4 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.