Jonas Le Fevre Sejersen
Robotic Software Engineer at BEUMER Group
Aarhus, Central Denmark Region
Summary
👤
Senior🎓
Top SchoolJonas Le Fevre Sejersen is a robotic software engineer and early-career Ph.D. candidate specializing in multi-agent deep learning for long-term autonomy, with ten years of industry and research experience building real-world multi-robot systems. He combines practical skills in Python, C++, ROS, Unity and Linux with hands-on work creating digital twins, large-scale image processing, object detection models and 3D visualizations for industrial automation. His research explores how graph neural networks and learning-based decision-making can coordinate collaboration among fleets of robots, and he prototypes algorithms in simulated environments like AirSim/Unreal to bridge lab results to deployment. Based in Aarhus, Denmark, Jonas has moved between applied R&D and product engineering at BEUMER Group, giving him a rare perspective on both academic rigor and production constraints. Colleagues value his focus on scalable, visualizable solutions and his habit of implementing state-of-the-art GNN models to test what’s practically achievable today.
10 years of coding experience
6 years of employment as a software developer
Master's degree in Computer Science, Deep learning, Computer Vision and Robotics, Master's degree in Computer Science, Deep learning, Computer Vision and Robotics at Aarhus UniversityNanodegree in Deep Reinforcement Learning, Computer Software Engineering, Nanodegree in Deep Reinforcement Learning, Computer Software Engineering at UdacityMatematik, Fysik, kemi, Matematik, Fysik, kemi at Silkeborg Teknisk Gymnasium
Danish, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (100)
simulator10
unreal10
graph-convolutional-networks10
robotics10
control-systems10
deep-reinforcement-learning10
computer-vision10
airsim10
platform-independent10
graph10
graph-neural-network10
simulation10
geometric-deep-learning10
pytorch10
robot10
Programming languages (8)
Github contributions (5)
Zartris/VectorizedMultiAgentSimulator
May 2023 - Jul 2024
VMAS is a vectorized framework designed for efficient Multi-Agent Reinforcement Learning benchmarking. It is comprised of a vectorized 2D physics engine written in PyTorch and a set of challenging multi-robot scenarios. Additional scenarios can be implemented through a simple and modular interface.
Contributions:2 PRs, 53 pushes, 11 branches in 1 year 2 months
Zartris/DRL-project
Oct 2019 - Jan 2020
Contributions:81 commits, 86 pushes, 1 branch in 3 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.