39 open roles in General AI Evaluation

What the work involves

Sharp generalists rate, compare and stress-test AI across domains.

• Compare and rank AI responses (preference data).
• Rate accuracy, helpfulness and safety.
• Red-team models to find failure modes.
• Write and refine prompts and test cases.

Who fits

Detail-oriented people with strong judgment from any background. No AI background required — your domain expertise is the qualification.