Data-Driven Robotics Research Engineer – Foundation Models

Innodata Inc.; New Jersey; Hybrid; USD 180,000 - 300,000; Full time

Job summary

Innodata Inc. is seeking a senior- to mid-level robotics ML specialist to shape data for foundation models in robotics. You will translate model requirements into concrete data specs, guide data collection across modalities, and run experiments to demonstrate data-driven model improvements.

Collaboration with labs and research teams is essential. Strong Python, PyTorch, and a data-centric mindset are required, with experience in real-world robotics data formats and simulated environments.

Qualifications

• Background in robot learning or robotics machine learning.
• Hands-on experience training and evaluating models for robotics with PyTorch.
• Familiar with robotics data formats and standards.
• Data-centric mindset; data drives model performance.
• Strong Python and reproducible experiments.
• Experience fine-tuning large vision-language models with modern toolchains.
• Experience with teleoperation or egocentric data collection.
• Publications or open-source contributions in robot learning.

Responsibilities

• Translate robotics model requirements into concrete data specifications for modalities and annotations.
• Guide data structuring, formatting, and enrichment for robot foundation models.
• Direct data collection across capture modalities to map to model needs.
• Run experiments to validate data design through fine-tuning and evaluation.
• Develop benchmarks for robotics data and model evaluation.
• Collaborate with capture lab, annotation teams, and synthetic data pipelines.
• Represent Innodata’s data approach in technical talks with researchers and robotics groups.

Skills

robotics ML

PyTorch fundamentals

Python

communication skills

data-centric mindset

Tools

NVIDIA Isaac Sim

Omniverse

Isaac Lab

Job description

Innodata Inc. is seeking a senior- to mid-level robotics ML specialist to shape data for foundation models in robotics. You will translate model requirements into concrete data specs, guide data collection across modalities, and run experiments to demonstrate data-driven model improvements.

Collaboration with labs and research teams is essential. Strong Python, PyTorch, and a data-centric mindset are required, with experience in real-world robotics data formats and simulated environments.