Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will conduct original research on social world models, vision-language-action architectures, and embodied reasoning, taking work from initial ideas through deployment. You will develop data strategies for training and evaluation, build rigorous simulation and real-world benchmarks, publish research, and create production-quality research code for learned robotic policies.
Requirements
- MS or PhD in machine learning, robotics, or a related field, or equivalent research experience with a strong publication record
- Deep expertise in world models, VLA models, imitation learning, reinforcement learning, or multimodal foundation models
- Strong engineering fundamentals in Python, Go, and/or C++
- Ability to train models and write production-quality research code
Responsibilities
- Conduct original research on social world models, VLA architectures, and embodied reasoning
- Drive research from idea to deployed system
- Design data strategies for training and evaluation using real-world deployment data
- Build rigorous evaluations for socially intelligent behavior
- Publish at leading research venues and represent the organization in the research community
- Collaborate to deploy research on physical robots