- 经验
- 任何
- 薪水
- 每小时 20 至 22 美元
- 职位空缺
- 1
- 发布
- 1天前
- 工作模式
- 在家办公
- 合格
- Candidates with relevant experience in AI adversarial work, cybersecurity, or socio-technical testing who can work remotely and commit 10 to 40 hours per week are suitable.
- 恢复
- 需要申请
职位描述
Role Overview
This contract role is for AI safety specialists who can evaluate conversational AI systems in English and Bengali. The work is fully remote and is suited to people who can commit between 10 and 40 hours each week. Compensation is set at $20 to $22 per hour.
What You’ll Do
- Test chatbots and AI agents by attempting jailbreaks, prompt injection attacks, and bias-triggering scenarios.
- Create well-labeled human review data by identifying failure modes, categorizing vulnerabilities, and highlighting broader safety risks.
- Use defined taxonomies, evaluation benchmarks, and testing playbooks so assessments stay consistent and repeatable.
- Prepare clear, reproducible documentation such as reports, datasets, and attack examples for client use.
What We’re Looking For
- Previous experience in AI red teaming, adversarial testing, cybersecurity, or related socio-technical analysis.
- A naturally investigative, adversarial approach with the ability to probe systems until weaknesses surface.
- A disciplined, framework-driven method rather than ad hoc experimentation.
- Strong written and verbal communication skills for explaining risks to both technical and non-technical audiences.
- Flexibility to move between different projects and client needs.
Application Steps
- Submit your resume.
- Complete an interview.
- Fill out the required form.