Karl Pertsch
Member Of Technical Staff at Physical Intelligence
Philadelphia, Pennsylvania, United States
Summary
🤩
Rockstar🎓
Top SchoolKarl Pertsch is a Member of Technical Staff and researcher with a decade of experience at the intersection of computer vision, robotics, and machine learning, currently splitting time between Physical Intelligence and postdoctoral roles at Stanford and UC Berkeley. He holds a PhD in Computer Science from USC and built a strong foundation in 6D pose estimation, RGB/RGB‑D reconstruction, and reinforcement-learned control through projects at TU Dresden, Penn GRASP, Berkeley RAIL, and industry internships at BMW. Karl’s work spans both academic research and applied systems—designing perception pipelines robust to heavy occlusion and energy-optimized driving strategies—making him adept at moving novel algorithms toward real-world robotics applications. Based in Philadelphia, he combines rigorous engineering (Dipl.-Ing. with a 3.95 GPA) with a track record of cross-institutional collaboration that often bridges simulation, ROS-based deployment, and lab-to-field transition.
10 years of coding experience
6 years of employment as a software developer
PhD, Computer Science, PhD, Computer Science at University of Southern CaliforniaDiplom-Ingenieurin (Dipl.Ing.), Electrical and Electronics Engineering, GPA: 3.95, Diplom-Ingenieurin (Dipl.Ing.), Electrical and Electronics Engineering, GPA: 3.95 at Technische Universität DresdenComputer Science, Computer Science at University of Pennsylvania
English, French, Japanese
Github Skills (25)
reinforcement-learning10
robotics9
datasets9
machine-learning7
tensorflow7
transformer7
robotics-simulation7
jax6
simulation6
hierarchical6
computer-vision5
numpy5
horizon4
vision-and-language4
robot4
Programming languages (5)
Github contributions (5)
kpertsch/d4rl
Apr 2020 - Mar 2022
A benchmark for offline reinforcement learning.
Contributions:77 commits, 1 PR, 58 pushes in 1 year 10 months
reinforcement-learning
clvrai/spirl
Oct 2020 - Apr 2022
Official implementation of "Accelerating Reinforcement Learning with Learned Skill Priors", Pertsch et al., CoRL 2020
Contributions:45 commits, 10 PRs, 50 pushes in 1 year 6 months
reinforcement-learningrobot-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.