Researcher

The role

A researcher will work on core problems in robot learning and data quality. You will design experiments on training robotics, vision, and embodied AI models using real-world data.

What you will do

• Design and run experiments on training robotics, vision, and embodied AI models using Trace's real-world data.
• Investigate how data quality, structure, and diversity affect downstream model performance.
• Turn findings into concrete recommendations for what to capture.
• Build and iterate on evaluation methodology.
• Stay close to the literature on robot learning and translate relevant work into what can be shipped.
• Work directly with the engineering and data teams to turn research findings into pipeline and product changes.

Required qualifications

• Currently completing, or recently completed, a Master's or PhD program with a research focus on robot learning, robotics, or training foundation/AI models.
• Exceptional undergraduates from top-tier robotics programs will also be considered.
• If coming from industry, background should be a researcher role at a robotics company or an equally reputable research lab.
• Motivated by seeing research translate into working systems.
• Optimistic and serious about where robotics and AI are headed.
• Comfortable being early in your career but still forming strong, defensible technical opinions.

Nice to have

• Publications or research work specifically in robot learning or training models for physical/embodied systems
• Experience from a leading robotics research program or lab
• Hands-on exposure to real robotic hardware or real-world (versus purely simulated) data

About the company

Trace is building the data infrastructure that lets robots learn from the real world. They build infrastructure to capture and transform real-world data into training data for robots and other intelligent systems.