XDOF

Data pipelines and annotation for robotics

Vision/Language Research Scientist

AI ResearchConfirmed live in the last 24 hours

Full-timeSan Mateo, CA, USA$200k - $330k /yrHybrid

About the Job

Requirements

A PhD in Computer Science, Robotics, Machine Learning, or a related field from a top-tier program is required.
Three to seven years of research or applied research experience in industry or academia in one or more of vision-language models, video understanding, robot learning, or generative modeling is required.
Deep fluency in PyTorch and working knowledge of large-scale training infrastructure, including distributed training, mixed precision, and large-batch workflows, are required.
Published work or demonstrable impact in vision-language models or vision-language-action models, video representation learning, imitation learning, or a closely related area is required.
Strong engineering fundamentals, including the ability to design clean systems rather than only run experiments, are required.

Responsibilities

Lead technical efforts at the intersection of vision-language models and robot learning.
Lead an exploratory project investigating what data a model needs to reason about and act in the physical world.
Study how frontier models fail at embodied tasks, form hypotheses about structured annotations on egocentric video that could address those gaps, prototype the annotations on the company’s data, and run experiments to evaluate their effectiveness.
Develop and evaluate representations that bridge visual perception, language, and low-level robot action conditioned on different modalities.
Collaborate directly with partner labs to define data requirements and connect data quality with downstream policy performance.
Stay current on research in vision-language-action models, video foundation models, flow matching, DiT architectures, and egocentric pretraining, and translate insights into new research and production systems.

Frontier labs are racing to build general-purpose robots, and the bottleneck isn't compute. It's data. At xdof, we're building the foundation behind the foundation models: the data collection systems, annotation pipelines, exabyte-scale data infrastructure, and software toolchain that enable our partners to push the field forward.

We're hiring a Research Scientist to lead technical efforts at the intersection of vision-language models and robot learning. You will lead an exploratory project on what data a model actually needs in order to reason about and act in the physical world. This is an open question across the field, and the role is largely about finding the answer: studying how frontier models currently fail at embodied tasks, forming hypotheses about which kinds of structured annotations on egocentric video would close those gaps, prototyping them on our data, and running experiments that show whether they help.

The ideal candidate has a research background in representation learning, and some of the following: vision, robotics, and language models.

What You'll Do

Develop and evaluate representations that bridge visual perception, language, and low-level robot action conditioned on different modalities Collaborate directly with partner labs to define data requirements and close the loop between data quality and downstream policy performance
Stay current on the research frontier (VLAs, video foundation models, flow matching, DiT architectures, egocentric pretraining) and translate insights into new research and production systems

About You

Required:

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 — you can design clean systems, not just run experiments

About the Company

XDOF

Data pipelines and annotation for robotics

XDOF builds data pipelines, data-collection tools, and annotation systems that frontier AI labs and robotics companies use to train robots to operate in the physical world. It tackles the data bottleneck in physical AI by combining teleoperation, simulation, and egocentric data collection, then layering in data cleaning, tooling, and annotation to prepare usable datasets. The company differentiates itself through its end-to-end data infrastructure for physical AI, its mix of teleoperation, simulation, and first-person data methods, and its Berkeley-founded background from 2024. Its goal is to enable robots to learn to operate in the real world by providing the pipelines and labels needed for training.

Company size: 1-10
Company stage: Series A
Total funding: $78.3M
Headquarters: Berkeley, CA
Founded: 2024

Recently Posted Jobs

Explore more opportunities at XDOF

See more jobs