仕事内容

Role overview

This is a remote contract opportunity for AI Safety Experts fluent in English and Punjabi. The work focuses on evaluating conversational AI systems and agents for safety issues, with pay set at $20 to $22 per hour. The expected weekly commitment ranges from 10 to 40 hours.

What you will do

Probe conversational AI models and agents for weaknesses by testing jailbreak paths, prompt injection attacks, and bias-related failures.
Create labeled human data by reviewing failures, classifying different types of vulnerabilities, and identifying broader systemic risks.
Use defined taxonomies, benchmarks, and testing playbooks to keep evaluations consistent and repeatable.
Capture results in a clear, reproducible format through reports, datasets, and concrete attack examples that can be used by customers.

What we are looking for

Hands-on background in AI red teaming, adversarial testing, cybersecurity, or socio-technical analysis.
A naturally investigative and adversarial approach that helps uncover edge cases and failure modes.
A disciplined, framework-based testing style rather than ad hoc experimentation.
Ability to communicate risk findings effectively to both technical teams and non-technical audiences.
Comfort working across different projects and client contexts with flexibility and adaptability.

Application process

Submit a resume.
Participate in an interview.
Complete the application form.

Additional information

This role is a contract position with remote work arrangements. Compensation is hourly at $20 to $22. The expected workload is 10 to 40 hours per week. The language requirement specified for the position is English and Punjabi.

Content Safety Analyst