SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
xDOF is building the foundational data infrastructure behind frontier robotics. The company focuses on data collection systems, annotation pipelines, exabyte-scale data infrastructure, and software toolchains that enable partners to advance general-purpose robot development.
You will lead technical efforts at the intersection of vision-language models and robot learning. This role centers on an exploratory research project investigating what data models actually need to reason about and act in the physical world—a core open question in the field. Your work will involve studying how frontier models currently fail at embodied tasks, forming hypotheses about which structured annotations on egocentric video could close performance gaps, prototyping annotations on the company's data, and running experiments to validate whether they improve downstream performance.
Key responsibilities include:
- Developing and evaluating representations that bridge visual perception, language, and low-level robot action across different modalities
- Collaborating directly with partner labs to define data requirements and close the feedback loop between data quality and downstream policy performance
- Staying current on the research frontier (VLAs, video foundation models, flow matching, DiT architectures, egocentric pretraining) and translating insights into new research and production systems
The ideal candidate brings a research background in representation learning, with experience in vision, robotics, and/or language models.
REQUIREMENTS:
- PhD in Computer Science, Robotics, Machine Learning, or a related field from a top-tier program
- 3–7 years of research or applied research experience (industry or academic) in one or more of: vision-language models, video understanding, robot learning, or generative modeling
- Deep fluency in PyTorch; working knowledge of large-scale training infrastructure (distributed training, mixed precision, large batch workflows)
- Published work or demonstrable impact in VLMs/VLAs, video representation learning, imitation learning, or a closely related area
- Strong engineering fundamentals—ability to design clean systems, not just run experiments
