Summary
Adrian Llopart is a founder and senior machine learning researcher with 10 years’ experience building production-ready deep learning systems for computer vision and robotics. He combines a PhD in Robotics/AI with hands-on expertise in instance/semantic segmentation, 3D human keypoint estimation, action recognition, SLAM and point-cloud processing to bridge research and deployed edge/cloud inference. At Veo and Huawei he led end-to-end pipelines—data collection, model design, optimization and inference infrastructure using AWS and GStreamer—while supervising students and shipping real-time systems. His work spans humanoid and service robotics, ROS integration, and manipulation tasks, reflecting a rare mix of control-theory grounding and modern deep learning. Now leading MederiAI in Copenhagen, he focuses on translating advanced perception models into robust products for human-centric environments. Colleagues describe him as equally comfortable prototyping novel algorithms and hardening them for constrained embedded and cloud deployments.
10 years of coding experience
6 years of employment as a software developer
BEST Course, Biomedical Signal Processing, BEST Course, Biomedical Signal Processing at Universidade Nova de Lisboa
BEST Course, Advanced Robotics, BEST Course, Advanced Robotics at Politehnica University of Timisoara
UPC Universitat Politècnica de Catalunya
Technical University of Denmark
University of Tokyo
Visiting Research PhD, Mechatronics, Robotics, and Automation Engineering, Visiting Research PhD, Mechatronics, Robotics, and Automation Engineering at Korea Advanced Institute of Science and Technology
Spanish, English, German, Catalan