Senior AI Safety Red Team Consultant (Remote)

Moonshot; Greater London; Remote; GBP 148,000 - 221,000; Full time

Benefits offered by this job

Flexible working arrangements

Diverse, impactful projects

Competitive consultancy rates

Remote working options available

Job summary

Moonshot is seeking a Senior Freelance Consultant to support the delivery of our AI Safety portfolio. This hands-on, technical role focuses on red teaming, adversarial evaluation, and methodological development across harm categories including CSEA, extremism, crisis, and teen safety.

You will translate violence prevention and safeguarding expertise into structured, testable outputs for technical and policy audiences, working within an ethical, legally compliant framework.

Qualifications

• Experience in trust and safety, online harms, or related fields and applying that knowledge to AI systems.
• Designing research, evaluation frameworks, or interventions for harm categories like violent extremism or CSEA.
• Translate real world harm knowledge into testable AI safety tests.
• Comfort working with highly sensitive or graphic content; wellbeing practices understood.

Responsibilities

• Red team and adversarial evaluation of AI systems against defined harm categories.
• Review model responses against harm criteria and provide expert judgement.
• Apply subject matter expertise to harm areas such as grooming, radicalisation, crisis signalling, or teen safety.
• Design evaluation frameworks translating real world harm knowledge into structured, testable criteria.
• Contribute to intervention logic connecting at risk users to support services.
• Draft methodology or findings for technical and government audiences.

Skills

Trust & safety experience

AI safety / online harms

Ability to test AI systems for harm

Strong written communication

Ability to work with sensitive content

Security clearance willingness

Job description

Moonshot is seeking a Senior Freelance Consultant to support the delivery of our AI Safety portfolio. This hands-on, technical role focuses on red teaming, adversarial evaluation, and methodological development across harm categories including CSEA, extremism, crisis, and teen safety.

You will translate violence prevention and safeguarding expertise into structured, testable outputs for technical and policy audiences, working within an ethical, legally compliant framework.