Developing advanced AI capabilities for robotics and building the foundational AI ecosystem.
Requirements
- Proficiency in multimodal models, modern ML architectures (e.g., transformers, diffusion models), and reinforcement learning.
- Proficiency in deep learning frameworks such as PyTorch, JAX, or TensorFlow.
- Exposure to robot learning through tactile and/or vision-based sensors.
- Expertise in object pose estimation, bimanual manipulation, and synthetic data.
Responsibilities
- Research and experiment with new ML models for perception and/or complex manipulation tasks, including: Spatial AI: 3D reconstruction, neural rendering, SLAM, etc.
- State Estimation: pose estimation, tracking, depth estimation, etc.
- Robot Learning: imitation learning, reinforcement learning, etc.
- Multimodal Fusion: model combining different sensor inputs and modalities.
- Collaborate with robotics, hardware, and product teams to ensure seamless AI model integration.
- Conduct experiments and evaluations with the goal of advancing state-of-the-art performance and contributing to research publications.
Other
- Currently pursuing or recently completed a PhD degree in Computer Science, Robotics, Machine Learning, or a related field.
- Requires 5 days/week in-office collaboration with the teams.
