Refusal Preference Preference Data Reviewer

AI Trainer Jobs; United States; Remote; USD 90,000 - 96,000; Full time

Job summary

AuraOne is seeking a contractor to perform a Refusal Preference Preference Data Reviewer role. Remote work within the US is available, with hourly compensation and responsibilities focused on adversarial evaluation and documentation.

You'll design jailbreak scenarios, document failures with rigor, and help patch gaps before models ship. Expect to operate as an independent specialist with strong reporting and communication skills.

Qualifications

• Background in offensive security, AppSec, or trust & safety operations.
• Experience publishing or reproducing public adversarial-ML research.
• Multilingual fluency for cross-language attack testing.

Responsibilities

• Design adversarial prompts that probe known weakness classes for Refusal Preference Preference Data Reviewer assignments.
• Document every successful attack with reproduction steps and the policy clause it violated.
• Score model defenses across single-turn and multi-turn conversations.
• Triage emerging attack vectors and route them to the safety team with severity ratings.

Skills

Adversarial prompting

Red-team analysis

Policy taxonomy

Failure documentation

Refusal Preference Data evaluation

Preference ranking

RLHF

Rater calibration

Refusal

Preference

Job description

Refusal Preference Preference Data Reviewer is a remote red-team track for stress-testing AI systems against adversarial prompts. Reviewers craft attack scenarios, document the failure mode, and pair each successful jailbreak with the rubric clause it violated so the safety team can patch the gap.

Category: AI Safety & Red Teaming · Pay: $65–$70 / hr · Location: Remote — US-eligible · Contractor

Refusal Preference Preference Data Reviewer is a remote red-team track for stress-testing AI systems against adversarial prompts.

About the role

Refusal Preference Preference Data Reviewer is a remote red-team track for stress-testing AI systems against adversarial prompts. Reviewers craft attack scenarios, document the failure mode, and pair each successful jailbreak with the rubric clause it violated so the safety team can patch the gap.
Adversarial evaluation is how AuraOne hardens AI models before they ship to customers. Reviewers think like attackers and write up failures with enough rigor that the modeling team can reproduce, fix, and regress-test them.
Push on refusal boundaries and dual-use risk before a model ships.

Responsibilities

• Design adversarial prompts that probe known weakness classes (jailbreak, policy bypass, prompt injection) for Refusal Preference Preference Data Reviewer assignments.
• Document every successful attack with reproduction steps and the policy clause it violated.
• Score model defenses across single-turn and multi-turn conversations.
• Triage emerging attack vectors and route them to the safety team with severity ratings.

Role details

Track Adversarial evaluation Work model Remote · Independent specialist contractor Compensation Hourly rate confirmed after the interview process. Eligible from US

What you should bring

• Demonstrated experience red-teaming AI systems, security research, or adversarial ML work for Refusal Preference Preference Data Reviewer work.
• Strong written communication — your reports become the patch ticket.
• Comfort working in policy-grey areas with clear documentation of what was attempted and why.
• Familiarity with prompt-injection, jailbreak, and policy-bypass taxonomies.
• Reliable async availability for at least 10 hours per week.

Role signals

Example tasks

• Construct a 5-turn adversarial conversation that bypasses a specific policy clause and write up the patch ticket.
• Score a model's defenses against a known jailbreak pattern across 20 variants.
• Propose a new red-team rubric category after spotting an emerging attack vector.
• Reproduce a failure another reviewer reported and confirm the severity tag.

Useful experience

• Background in offensive security, AppSec, or trust & safety operations.
• Experience publishing or reproducing public adversarial-ML research.
• Multilingual fluency for cross-language attack testing.

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

• Adversarial prompting
• Red-team analysis
• Policy taxonomy
• Failure documentation
• Refusal Preference Preference Data evaluation
• Preference ranking
• RLHF
• Rater calibration
• Refusal
• Preference

Application boundary

Creating a specialist profile records your experience and preferences. Starting role intake is a separate action that attaches this role to your candidate record.

Specialist intake

The intake preserves your chosen role, the visible terms, and source attribution for reviewer context.

• 01 Confirm profile and eligibility details.
• 02 Attach this role deliberately.
• 03 Receive a human review decision or follow-up.

Placement timing depends on program demand and reviewer confirmation.