Robotics Data Trainer

AI Trainer Jobs; United States; Remote; USD 34,000 - 83,000; Full time

Job summary

AuraOne is seeking a Robotics Data Trainer for a remote review track evaluating robotics outputs against real-world physical-AI constraints. Reviewers grade trajectory choices, perception assumptions, and safety envelopes; flag failure modes; and document corrected behavior so the modeling team can train on it.

Responsibilities include reviewing outputs against safety standards, scoring perception and motion, and adjudicating cases using firm operating procedures.

Qualifications

• Hands-on experience with robotic, autonomous, or teleoperated systems.
• Ability to reason about perception, motion planning, and safety envelopes.
• Familiarity with structured incident taxonomies and severity scoring.
• Clear written explanations naming failure modes and the standards being applied.
• Reliable async availability for at least 10 hours per week.

Responsibilities

• Review robotics outputs against current safety standards and operating envelopes for Robotics Data Trainer assignments.
• Grade perception, motion, and decision logic on a structured rubric.
• Flag near-miss and unsafe trajectories with severity tags that match incident taxonomies.
• Capture the corrected behavior or reference trajectory so the modeling team can train on it.
• Adjudicate disputed cases against published standards or firm operating procedures.

Skills

Robotics experience

Perception reasoning

Safety envelopes

Incident taxonomy

Written reasoning

Async availability

Job description

Robotics Data Trainer is a remote review track for evaluating robotics review outputs against real-world physical-AI constraints. Reviewers grade trajectory choices, perception assumptions, and safety envelopes; flag failure modes; and document the corrected behavior so the modeling team can train on it.

Category: Robotics & Embodied AI · Pay: Hourly rate confirmed after the interview process · Location: Remote — US‑eligible · Contractor

Robotics Data Trainer is a remote review track for evaluating robotics review outputs against real-world physical-AI constraints.

Role details

About the role

Robotics Data Trainer is a remote review track for evaluating robotics review outputs against real-world physical-AI constraints. Reviewers grade trajectory choices, perception assumptions, and safety envelopes; flag failure modes; and document the corrected behavior so the modeling team can train on it.
Robotics review systems run in the physical world, where a confident-looking failure can hurt someone. AuraOne uses experienced robotics and operations reviewers to grade outputs against perception, motion, and safety constraints.
Review teleoperation, navigation, and manipulation work. Physical-world safety is part of it.

Responsibilities

• Review robotics review outputs against current safety standards and operating envelopes for Robotics Data Trainer assignments.
• Grade perception, motion, and decision logic on a structured rubric.
• Flag near-miss and unsafe trajectories with severity tags that match incident taxonomies.
• Capture the corrected behavior or reference trajectory so the modeling team can train on it.
• Adjudicate disputed cases against published standards or firm operating procedures.

What you should bring

• Hands-on experience with robotic, autonomous, or teleoperated systems. Operating, supervising, or evaluating them for Robotics Data Trainer work.
• Comfort reasoning about perception, motion planning, and physical safety envelopes.
• Familiarity with structured incident taxonomies and severity scoring.
• Clear written reasoning that names the failure mode and the standard being applied.
• Reliable async availability for at least 10 hours per week.

Role signals

Example tasks

• Grade a robotics review model's chosen trajectory and flag any near-miss or unsafe motion.
• Score a perception output against ground-truth labels and rate the failure category.
• Adjudicate a disputed safety call between two reviewers using firm operating procedures.
• Audit a 25-trajectory batch for rubric consistency and report drift to the program lead.

Useful experience

• Background in robotics engineering, autonomy operations, or industrial safety.
• Prior experience reviewing AI-assisted robotics tooling and its failure modes.
• Familiarity with simulation environments and trajectory-replay tooling.

Compensation and schedule

Hourly rate confirmed after the interview process.
Expected arrangement: contractor , with program-defined task volume and review pacing. Placement depends on current program demand and reviewer confirmation.

Skills used in matching

• Robotics evaluation
• Perception review
• Motion safety
• Incident taxonomy
• Robotics review
• AI model evaluation