Safety Researcher (AI Safety)

Location: San Francisco office, on-site

Company

hirify.global works on AI systems whose behavior requires safety research and evaluation.

What you will do

• Build evaluations and investigate concerning model behavior.
• Assess whether proposed work should proceed and raise safety questions in meetings.
• Document initial observations and develop written records of safety concerns.
• Communicate concerns calmly in meetings and in writing.

Requirements

• Strong research background in AI safety, alignment, interpretability, or model evaluation.
• Experience raising concerns calmly and documenting them in writing.
• Ability to handle sensitive information and communicate when specifics cannot be disclosed.
• Sincere interest in improving AI safety from inside an organization.

Culture & Benefits

• Contract position with a defined six-week engagement.
• Work directly with highly talented colleagues.
• The role involves access to model behavior and internal safety investigations.

Hiring process

• Submit a draft of a potential resignation post; specific details may be omitted.