AuraOne is seeking a remote contractor to perform Robot Policy Compliance Task evaluation and adversarial testing of AI systems. You will craft 5-turn jailbreak scenarios, document failures, and map each to the corresponding policy clause to help patch gaps before release.
Ideal candidates have red-teaming or adversarial ML experience and can document findings clearly for the modeling team to reproduce, fix, and regression-test.
#J-18808-Ljbffr