Abdul Kreidieh
Software Engineer at YouTube
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolAbdul Kreidieh is a software engineer and PhD-trained researcher who applies modern machine learning and reinforcement learning to real-world mobility and geospatial problems. With nine years of experience spanning UC Berkeley, Google Research, and YouTube, he has built simulation platforms for mixed-autonomy traffic, contributed to a major open-source RL traffic framework (flow), and helped run the first large-scale field demo showing automated vehicles can reduce congestion. He specializes in adapting ML algorithms to novel domains—combining hierarchical RL, geospatial reasoning, and statistical policy evaluation—and enjoys interdisciplinary teams that bridge research and production. Notably, his work has influenced both academic experiments and product-level systems, from congestion pricing studies to recommender-system RL at scale.
9 years of coding experience
9 years of employment as a software developer
Doctor of Philosophy - PhD, Civil Engineering, Doctor of Philosophy - PhD, Civil Engineering at University of California, BerkeleyMaster of Science - MS, Civil Engineering, Master of Science - MS, Civil Engineering at UC Berkeley College of EngineeringLebanese Baccalaureate: Official in 2012, Mechanical Engineering, Lebanese Baccalaureate: Official in 2012, Mechanical Engineering at American Community School BeirutBachelor of Engineering (B.E.), Mechanical Engineering, Bachelor of Engineering (B.E.), Mechanical Engineering at American University of Beirut
Arabic, English
Github Skills (4)
object-oriented-programming10
python10
data-structure9
data-structures9
Programming languages (4)
Github contributions (5)
flow-project/flow
Feb 2017 - Jun 2021
Computational framework for reinforcement learning in traffic control
Role in this project:
Back-end Developer
Back-end DeveloperContributions:5 releases, 3077 commits, 432 PRs in 4 years 4 months
Contributions summary:Abdul contributed to the development of the base scenario class by implementing default edge_starts and adding the functionality to specify route and connection information for a highway network. They also refactored vehicle data into the environment to support a wider range of functionalities. The user made additions to support a more modular and flexible routing algorithm.
reinforcement-learningbenchmarkautonomousvehicle-controlsumo
dissendahl/flow
Jul 2019 - Nov 2019
Computational framework for reinforcement learning in traffic control
Contributions:53 commits in 4 months
reinforcement-learning
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.