Posted 23 Sept 2026
AI Red Team Tester
We're looking for AI Red Team Testers to think like an adversary — probing, challenging, and stress-testing cutting-edge AI systems to expose weaknesses before they cause real-world harm.
Hiring companyAlignerr
About the role
What if your curiosity, creativity, and knack for finding loopholes could directly make AI safer for millions of people?
This is a fully remote, flexible contract role open to anyone who thinks critically, loves puzzles, and enjoys finding the cracks in seemingly solid systems. No cybersecurity or hacking background required.
Why Join Us
- Work on meaningful AI safety projects alongside leading research labs.
- Fully remote and flexible — work on your own schedule, wherever you are.
- Freelance autonomy with the satisfaction of genuinely impactful work.
- Contribute to making AI safer, fairer, and more reliable for everyone.
- Potential for ongoing work and contract extension as new projects launch.
Responsibilities
- Design creative, adversarial prompts and scenarios designed to expose AI weaknesses.
- Attempt to elicit incorrect, unsafe, biased, or inappropriate outputs from AI models.
- Document every failure mode you discover with clear, reproducible steps.
- Rate and categorize the severity of issues based on structured guidelines.
- Collaborate asynchronously with AI safety and research teams to report findings.
- Explore edge cases across a wide range of topics, tones, and scenarios.
Required skills
- A creative, curious thinker who enjoys puzzles, thought experiments, and finding flaws in systems.
- Comfortable thinking adversarially — you naturally ask "what could go wrong?"
- Detail-oriented and systematic: you document findings clearly and thoroughly.
- Strong written communication skills in English.
- Self-motivated and able to work independently on task-based assignments.
Preferred skills
- Experience in writing, journalism, creative fiction, or debate.
- Familiarity with AI language models as an end user.
- Background in ethics, philosophy, psychology, or critical analysis.
- Prior experience in QA, software testing, or content evaluation.
Experience and education
- No background in cybersecurity, hacking, or AI required.
Schedule details
10–40 hours per week.
Eligibility
Work arrangement: Fully remote
Eligible countries: Argentina
Compensation details
$40–$120 per hour.
More about the hiring companyAlignerr
