Remote AI Research Auditor - STEM Derivation & Validation

AuraOne; United States; On-site; USD 176,332 - 238,324; Full time

Job summary

AuraOne is seeking a STEM Professor contractor to remotely assess AI outputs across stem research reasoning, calculations, and workflows. You will review derivations, reproduce key steps, and document the correct method so the modeling team can train on it.

Responsibilities include evaluating model outputs against established methods, flagging dimensional, methodological, and citation errors with structured severity tags, and capturing worked examples for training.

Qualifications

• Graduate-level training or equivalent applied experience in STEM research or closely related field.
• Hands-on experience publishing, teaching, or advising on the topic at a professional level.
• Clear written reasoning citing methods, papers, or worked examples.
• Reliable async availability for at least 10 hours per week.

Responsibilities

• Review AI outputs against current STEM research methods, conventions, and prior work for STEM Professor assignments.
• Reproduce or sanity-check key derivations, calculations, or experimental claims.
• Flag dimensional, methodological, and citation errors with structured severity tags.
• Capture the corrected reasoning or worked example so the modeling team can train on it.
• Adjudicate disputed answers against textbooks, papers, or community standards.
• Maintain reviewer-quality scores in inter-rater calibration cycles.

Skills

Scientific reasoning

Method validation

Citation review

Quantitative analysis

STEM research

AI model evaluation

Education

PhD or equivalent

Job description

AuraOne is seeking a STEM Professor contractor to remotely assess AI outputs across stem research reasoning, calculations, and workflows. You will review derivations, reproduce key steps, and document the correct method so the modeling team can train on it.

Responsibilities include evaluating model outputs against established methods, flagging dimensional, methodological, and citation errors with structured severity tags, and capturing worked examples for training.