Robotics World-Models Research Scientist

Innodata Inc.; United States; Remote; USD 160,000 - 185,000; Full time

Job summary

Innodata is hiring a Research Scientist to own the science of data for robot foundation models in direct partnership with customers and frontier labs. You will define data designs, modalities, sampling, and evaluation criteria, weighing real-world capture against simulation and guiding multi-sensor data collection.

The role emphasizes experimental validation, sim-to-real transfer, and publishing benchmarks and methodologies that advance robotics research.

Qualifications

• Experience in robot learning or robotics ML.
• Strong understanding of data design for robotics foundation models.
• Experience with simulation and synthetic data pipelines.

Responsibilities

• Define how Innodata designs, structures, and evaluates data for robot foundation models and validate choices experimentally.
• Translate robotics model requirements into concrete data specifications and evaluation criteria.
• Decide what to capture in the real world vs. simulation and curate training mixes across diverse datasets.
• Guide capture modalities and multi-sensor data collection to align with model needs.
• Build evaluation benchmarks and publish methodologies to advance the field.

Skills

Robot learning

PyTorch fundamentals

Data curation

Sim-to-real transfer

Teleoperation data collection

Education

Bachelor's degree in a technical field

MS/PhD preferred

Tools

PyTorch

NVIDIA Isaac Sim

Omniverse

HuggingFace transformers

Job description

Innodata is hiring a Research Scientist to own the science of data for robot foundation models in direct partnership with customers and frontier labs. You will define data designs, modalities, sampling, and evaluation criteria, weighing real-world capture against simulation and guiding multi-sensor data collection.

The role emphasizes experimental validation, sim-to-real transfer, and publishing benchmarks and methodologies that advance robotics research.