Develop, train, fine-tune, and evaluate AI models for robotics and Physical AI applications.
Work with modern vision models , including object detection, segmentation, tracking, pose estimation, and vision foundation models.
Integrate and experiment with LLMs, VLMs, and multimodal models for robot perception, reasoning, planning, and task understanding.
Build data collection, preprocessing, training, evaluation, and inference pipelines.
Explore learning-based approaches for robot manipulation, including imitation learning, policy learning, and multimodal action models.
Evaluate new research models and quickly determine whether they can create practical value in our system.
Work closely with robotics and software engineers to bring AI models from experiments into real robot deployments.
Continuously explore new approaches in foundation models, embodied AI, and Physical AI.