AI Red Team Specialist

About the role

A contract role testing conversational AI models and agents for safety weaknesses. The specialist probes systems adversarially and documents findings so customers can act on them.

What you will do

• Red team conversational AI models and agents
• Attempt jailbreaks, prompt injections, misuse and multi-turn manipulation
• Annotate failures, classify vulnerabilities and flag systemic risks
• Apply taxonomies, benchmarks and playbooks to keep testing consistent
• Produce reproducible reports, datasets and attack cases
• Work independently and asynchronously while meeting deadlines

What they look for

• Native fluency in English and Punjabi
• Sound judgment about language and content
• Rigorous attention to detail and ability to spot subtle inconsistencies
• Structured approach and consistent adherence to quality standards
• Clear communication of reasoning to technical and non-technical audiences
• Adaptability across projects, task types and customers

Nice to have

• Adversarial machine learning experience such as jailbreak datasets, prompt injection or model extraction
• Cybersecurity background such as penetration testing or reverse engineering
• Experience probing socio-technical risks such as harassment or disinformation
• Creative probing skills from psychology, acting or writing

Pay

USD 16 to 22 per hour