ML Output Review Specialist (Remote)

AuraOne, Inc.; United States; Remote; USD 55,000 - 83,000; Full time

Job summary

AuraOne, Inc. is seeking an Applied Scientist — Machine Learning specialist operations contractor to review AI outputs and evaluate workflow adherence.

This remote role targets MSc or PhD holders, with US eligibility, to work on evaluating governance and operational risk across ML tasks. You will grade tone, escalation, and stakeholder fit using published playbooks, capture actionable next steps for the modeling team, and maintain calibration scores.

Qualifications

• Direct working experience in applied scientists machine learning specialist operations on real teams.
• Comfort applying multi-page rubrics consistently across long batches.
• Clear written reasoning that names the policy or workflow being applied.
• Reliable async availability for at least 10 hours per week.

Responsibilities

• Review AI outputs against current workflows, playbooks, and firm policy for Applied Scientists — Machine Learning (MSc. or PhD) assignments.
• Grade tone, escalation logic, and stakeholder fit on a structured rubric.
• Flag operational risk, missed escalations, and policy-adherence gaps with severity tags.
• Capture the right next step so the modeling team can train on it.
• Adjudicate disputed treatments against published playbooks or firm guidance.
• Maintain reviewer-quality scores in inter-rater calibration cycles.

Skills

Operational review

Policy adherence

Workflow judgment

Stakeholder communication

Applied Scientists Machine Learning

Education

MSc. or PhD in a relevant field

Job description

AuraOne, Inc. is seeking an Applied Scientist — Machine Learning specialist operations contractor to review AI outputs and evaluate workflow adherence.

This remote role targets MSc or PhD holders, with US eligibility, to work on evaluating governance and operational risk across ML tasks. You will grade tone, escalation, and stakeholder fit using published playbooks, capture actionable next steps for the modeling team, and maintain calibration scores.