Mercor is building a red team to test AI systems by probing models with adversarial inputs. This role focuses on reviewing AI outputs touching sensitive topics, with optional participation in higher-sensitivity projects and clear guidelines for wellbeing.
You will document reproducible artifacts, annotate failures, and generate attack cases that customers can act on, helping improve the safety and robustness of AI systems across projects and customers.
AI Safety Red Team Engineer (Remote) employer: Obsidian
At Obsidian, we pride ourselves on fostering a collaborative and innovative work culture that empowers our employees to excel in their fields. As a Finance & Real Estate Legal Expert, you will have the opportunity to engage in meaningful projects while benefiting from flexible working hours and a commitment to professional growth. Our supportive environment encourages continuous learning and offers unique insights into the intersection of law and technology, making it an ideal place for those looking to make a significant impact.
