Adversarial Prompt Expert
Half of all Red-Teaming & AI Safety roles come down within 21 days of being listed, and 19.3% are still up after four weeks. This one was posted 10 days ago.
About the role
The work
You will test multiple large language models to find unsafe, unexpected, or otherwise problematic behavior. The work is project-based, remote, and mostly asynchronous, with occasional strongly recommended synchronous meetings.
You will create adversarial prompts and test scenarios, assess model responses, and document findings clearly so they can support stronger AI defenses.
- Probe open-source and closed-source language models for failures and harmful outputs.
- Create jailbreak attempts, evasion tests, and edge-case scenarios across different topics.
- Review model responses and identify weaknesses in safeguards.
- Try varied approaches persistently, then record attempts and outcomes systematically.
- Research relevant topics with AI assistance and share documented findings with safety and engineering teams.
What it pays and takes
This is a flexible, part-time contractor role for candidates based in the United States.
- Pay: up to $45 to $65 per hour.
- Schedule: less than 20 hours per week, with an ideal commitment of at least 10 hours.
- Location: open to candidates based in the United States.
- Language: fluent in English.
- Experience: hands-on use of multiple open-source and closed-source language models.
- Skills: prompt engineering, including evasion techniques and jailbreaking; an adversarial or security mindset; persistence; creativity; careful documentation; and clear writing.
- Judgment: handle sensitive content responsibly and make ethical decisions.
- Helpful background: red teaming, offensive security, model evaluation, or comparable experience testing technical systems.
How it works
Apply on OpenTrain with your resume, then complete the application on the hiring site.
About AI training work
AI training is the human work behind systems that generate and understand language. People test model responses, find weaknesses, and provide clear feedback so AI systems can become more useful and safer. Experienced contributors are valued because careful testing helps uncover problems that automated checks may miss.
Where you can work
This role is 100% remote, for applicants located in United States.
From OpenTrain AI's job posting, checked 16 Nov 2026.
About OpenTrain AI
A unified board of AI-training jobs with real per-role pay.
How it pays: Stripe payouts to your bank, in the 120 countries its payout docs list.
When it pays: When the employer approves each funded milestone. How often others pay
Interview: on some roles or projects (AI interview (switched on per job by the employer)). How the OpenTrain AI AI interview works
Similar open roles
Some links on AITraining.jobs are referral links. If you sign up through one, AITraining.jobs may earn a commission at no extra cost to you. It never changes what a platform pays you. A role at a platform that pays us a referral fee weighs more in this order.

Open Source GitHub Maintainer Code Review (opens the job details)
OpenTrain AI · Coding · Remote· 1mo agoDetails
$120–$200/hr

Language Expert (Korean Audio Transcription) (opens the job details)
OpenTrain AI · Transcription & Audio · Remote· 1mo agoDetails
$20–$36/hr
Full-Stack Engineer (opens the job details)
OpenTrain AI · Coding · Remote· 2mo agoDetails
$50–$100/hr
AI Conversation Evaluator (opens the job details)
micro1 · AI Evaluation · Remote· 3d agoDetails
$40–$70/hr
Want companies to ask for you?
One profile for every AI training platform. Companies ask us for people with a specific language, country or credential: add yours once, and when a role fits, we introduce you, only with your yes. Free.
Get new OpenTrain AI roles by email
We check the OpenTrain AI job board every hour and email you its newest roles each week, plus similar paid roles elsewhere. Free, unsubscribe anytime.