## About Hub
Hub (Y Combinator 2026) collects real-world human and robot data to train robots, and works with some of the most recognised robotics and AI labs in the United States. Today Hub specialises in egocentric data collected at a scale and speed no lab could reach on its own: every contributor is identity-verified (KYC) and every contribution goes through several layers of quality control. Scalable QA of egocentric data, run by our Refinery team, remains a core company priority.
Hub Labs, headquartered in Toulouse, is where Hub turns that data into working robot behaviour. Within Hub Labs, the Brain team, led by Guillaume, works on robot learning across humanoids and other types of embodied machines.
## The role
You build the scientific program of Hub Labs with Guillaume. You start from what actually goes wrong when robots run, turn it into falsifiable questions and answer them with rigorous experiments. This role keeps real research capacity: time to think, design studies and publish, alongside hands-on work on physical robots.
**Location:** San Francisco, USA OR Toulouse, France. In-office and hands-on with physical robots; this is not a remote role.
**Location terms:** San Francisco-specific compensation and visa sponsorship arrangements are to be confirmed. The listed EUR range does not establish US employment terms.
## What you will do
- Formulate falsifiable research questions from robot failures observed in trials and deployments.
- Design rigorous experiments: hypotheses, controls, held-out evaluation and the data needed to answer them.
- Stay hands-on: write code, run experiments on real robots and read the results yourself.
- Work daily with robot learning and deployment engineers to turn findings into better data and better policies.
- Collaborate with academic peers, and publish and open-source work where it moves the field.
- Develop the lab's scientific program and research agenda with Guillaume.
- Mentor engineers and junior researchers.
## Who we are looking for
- A PhD in machine learning, robotics or a related field, followed by several years of research in academia or an industrial lab.
- Research depth in robot learning: publications at venues such as CoRL, RSS, ICRA, NeurIPS, ICML, ICLR or CVPR.
- Practical robot-learning evidence: policies you trained and evaluated on real robots, not only in simulation.
- Experience in areas such as imitation learning, VLAs, world models, RL, human-to-robot transfer or learning from video.
- A track record of experimental rigour and of working closely with engineers.
- Fluent English, written and spoken.
## Nice to have
- Experience with humanoids or other complex embodied systems.
- Work on egocentric video, teleoperation data or dataset and benchmark design.
- Open-source libraries, datasets or benchmarks used by other teams.
- Experience mentoring or co-supervising researchers.
## Why Hub Labs
- Real-world data at a scale few labs have, and the ability to collect new data when a question calls for it.
- A founding seat: the first researchers shape the lab's scientific direction.
- Your work goes straight into the data, models and benchmarks Hub builds with leading AI and robotics labs.
## How to apply
Send a short application, in your own words:
- Your CV
- The 2 or 3 papers or projects you are most proud of, including any work on physical robots
- Your Google Scholar, GitHub or website
- A few lines on a robot failure you would want to turn into a research question
Visa sponsorship is offered for this role.
