AI Safety Expert – English and Finnish (Remote)
Mercor is seeking English- and Finnish-speaking AI safety experts to red-team conversational AI models and agents. The work involves producing reproducible vulnerability data and attack cases for frontier AI training projects.
विवरण
- Conduct adversarial testing of conversational AI models and agents using jailbreaks, prompt injection, misuse scenarios, bias exploitation, and multi-turn manipulation
- Document failures, categorize vulnerabilities, and identify systemic risks
- Apply established taxonomies, benchmarks, and playbooks to maintain consistent evaluations
- Create reproducible reports, datasets, and attack cases
- Investigate model outputs involving bias, misinformation, and harmful behavior
- Broaden evaluation coverage to identify vulnerabilities that automated testing may overlook
- Work with project teams supporting the training and improvement of frontier AI systems
आवश्यकताएँ
- Fluent or native-level English and Finnish
- Previous red-teaming experience in AI adversarial testing, cybersecurity, or socio-technical probing
- Demonstrated ability to test AI systems adversarially
- Ability to work with frameworks, taxonomies, benchmarks, and playbooks
- Ability to communicate risks clearly to technical and non-technical audiences
- Adaptability across different projects and customers
- Must work as an independent contractor
- Must be able to work without confidential or proprietary information belonging to another employer, client, or institution
- H-1B and STEM OPT candidates are not supported
लाभ
- Fully remote contract role
- Flexible schedule with control over working hours
- Option to participate in higher-sensitivity projects
- Clear guidance and wellness resources
- Weekly payments through Stripe or Wise
- Competitive pay
- Referral opportunity offering up to $250 per successful referral
- Reasonable accommodations available upon request