AI Red Team Tester

Alignerr Corp.; Colombia; Remote; COP 88,353,000 - 176,706,000; Part time

Job summary

Alignerr is seeking AI Red Team Testers to think like adversaries and probe for weaknesses across AI systems. This is a fully remote, hourly contract offering flexible hours for 10–40 hours per week.

No cybersecurity background required; strong English writing and a curious, puzzle-loving mindset are key as you design prompts, document failures with reproducible steps, and collaborate with safety researchers. You will work across topics and testing methodologies to help build safer AI.

Qualifications

• A naturally curious thinker who enjoys puzzles, edge cases, and thinking outside the box.
• Comfortable approaching problems from an adversarial or contrarian angle
• Systematic and detail-oriented — you document what you find clearly and thoroughly
• Strong written communication skills in English
• Self-motivated and reliable in an async, remote work environment
• No hacking, cybersecurity, or AI background required

Responsibilities

• Design creative prompts and scenarios specifically intended to expose AI weaknesses
• Attempt to elicit incorrect, biased, harmful, or policy-violating outputs from AI models
• Document every failure you find with clear, reproducible steps
• Rate the severity and potential impact of each issue discovered
• Provide structured, detailed feedback using provided evaluation frameworks
• Collaborate asynchronously with AI safety and research teams
• Work across a range of AI systems, topics, and testing methodologies

Skills

Creative thinking

Adversarial mindset

Attention to detail

Strong English writing

Job description

About The Role

What if your curiosity and creative thinking could make AI systems safer for millions of people? We're looking for AI Red Team Testers to do something genuinely unique: try to break AI.

• Organization: Alignerr
• Type: Hourly Contract
• Location: Remote
• Commitment: 10–40 hours/week

About The Role

What if your curiosity and creative thinking could make AI systems safer for millions of people? We're looking for AI Red Team Testers to do something genuinely unique: try to break AI. Your job is to think like an adversary — crafting clever prompts, probing for weaknesses, and uncovering the failure modes that researchers need to know about. Every vulnerability you find helps build smarter, safer, and more reliable AI systems. No cybersecurity or technical background required. Just a sharp mind, a creative streak, and a love of finding what others miss.

What You'll Do

• Design creative prompts and scenarios specifically intended to expose AI weaknesses
• Attempt to elicit incorrect, biased, harmful, or policy-violating outputs from AI models
• Document every failure you find with clear, reproducible steps
• Rate the severity and potential impact of each issue discovered
• Provide structured, detailed feedback using provided evaluation frameworks
• Collaborate asynchronously with AI safety and research teams
• Work across a range of AI systems, topics, and testing methodologies

Who You Are

• A naturally curious thinker who enjoys puzzles, edge cases, and thinking outside the box
• Comfortable approaching problems from an adversarial or contrarian angle
• Systematic and detail-oriented — you document what you find clearly and thoroughly
• Strong written communication skills in English
• Self-motivated and reliable in an async, remote work environment
• No hacking, cybersecurity, or AI background required

Nice to Have

• Experience in creative writing, philosophy, logic, or debate
• Familiarity with AI chatbots or language models as an end user
• Background in research, journalism, or critical analysis
• Interest in AI ethics, safety, or policy

Why Join Us

• Work on cutting-edge AI safety projects alongside leading research labs
• Fully remote and flexible — work on your own schedule, wherever you are
• Freelance autonomy with meaningful, intellectually stimulating work
• Contribute to research that has a real impact on how AI behaves in the world
• Variety in every session — no two testing assignments are the same
• Potential for ongoing work and contract extension as new projects launch