Jan Blumenkamp
Research Associate
Cambridge, England, United Kingdom
Summary
👤
Senior🎓
Top SchoolJan Blumenkamp is a Research Associate and recent PhD graduate from the University of Cambridge with 11 years of engineering experience bridging embedded systems and multi-agent robotics. His research specializes in running graph neural network–based perception and control on real robot swarms, with prior work on emergent adversarial communication accepted to CoRL 2020. Jan pairs deep learning expertise with low-level embedded software experience from industry and research labs (including DLR and DFKI), enabling end-to-end deployments from FPGA and microcontroller firmware to distributed RL policies. Based in Cambridge, he focuses on practical transfer of GNN and MARL techniques into safety-conscious, resource-constrained robotic platforms. Colleagues describe him as someone who moves fluid research ideas into robust real-world systems, often by designing novel communication and redundancy mechanisms that work outside the lab.
12 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy - PhD, multi-agent robotics/deep learning, Doctor of Philosophy - PhD, multi-agent robotics/deep learning at University of CambridgeBachelor of Science - BS, Computer Science, Bachelor of Science - BS, Computer Science at University of Bremen
German, English
Github Skills (130)
parallel10
multi-agent-reinforcement-learning10
python10
simulator10
hyperparameter-optimization10
deep-learning10
gpu10
ray10
graph-convolutional-networks10
gpu-acceleration10
optimization10
robotics10
numpy10
llm-inference10
ai10
Programming languages (10)
Github contributions (5)
proroklab/ros2_multi_agent_passage
Dec 2021 - Mar 2022
Contributions:102 commits, 5 pushes in 2 months
proroklab/rllib_differentiable_comms
Jan 2021 - Jul 2022
This is a minimal example to demonstrate how multi-agent reinforcement learning with differentiable communication channels and centralized critics can be realized in RLLib. This example serves as a reference implementation and starting point for making RLLib more compatible with such architectures.
Contributions:37 commits, 15 pushes, 1 branch in 1 year 6 months
multi-agent-reinforcement-learningrllib
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.