AI Dataset Engineer for Robotic Systems
I designed and collected pick-and-place training datasets for robotic arms, focusing on high-variability manufacturing environments. I trained and evaluated models such as Pi0.5, GROOT, and ACT for embodied AI applications using these datasets. I researched and integrated AI solutions to improve robotic manipulation tasks for production systems. • Created custom dataset pipelines for robotic manipulation tasks • Evaluated multiple AI models in real-world scenarios • Focused on embodied AI within robotics and automation • Managed end-to-end data collection, labeling, and training processes