AI Safety Specialist (Red Teaming)

Responsibilities

• ▹Red team conversational AI models and agents for jailbreaks and prompt injections
• ▹Annotate failures and classify vulnerabilities
• ▹Apply structured testing using taxonomies and playbooks
• ▹Document reproducible reports and attack cases
• ▹Work independently and asynchronously to meet deadlines

Requirements

• ▹Native fluency in English and Finnish
• ▹Prior red teaming experience in AI adversarial work, cybersecurity or socio-technical testing
• ▹Strong communication skills for technical and non-technical stakeholders

Nice to have

• ▹Adversarial ML experience (jailbreak datasets, prompt injection, RLHF/DPO attacks)
• ▹Cybersecurity background (penetration testing, exploit development, reverse engineering)
• ▹Experience in socio-technical risk analysis

Soft skills

What we offer

• ▹Flexible, project-based contract work
• ▹Hourly rate of $48-62/hr

About the company

Mercor connects elite creative and technical talent with leading AI research labs; its investors include Benchmark, General Catalyst and Peter Thiel.