Senior AI Prompt Engineer, Trust Review Operations at LinkedIn3

Requirements

• ✓Design, evaluate, and iterate on prompts for generative and classification models supporting moderation, risk detection, case summarization, and reviewer assistance.
• ✓Conduct complex prompt-based investigations to diagnose model behavior, failure modes, and quality issues.
• ✓Establish frameworks for evaluating AI outputs across accuracy, bias, safety, and consistency.
• ✓Integrate AI tooling into Trust Operations workflows to support scalable and consistent resolution of high-risk cases.
• ✓Develop mitigation plans for model errors and propose long-term improvements that reduce operational and platform risk.
• ✓Serve as the Trust Operations escalation point for issues involving AI output quality, hallucinations, override rates, and misclassifications.
• ✓Ensure AI-generated outputs comply with platform policies, MDSS, safety standards, and international regulations (e.g., DSA, synthetic media rules).
• ✓Lead feedback-collection programs focused on AI output quality, partnering with Policy Operations, Data Science, Engineering, and human reviewers.
• ✓Analyze model and prompt performance data to generate insights and influence improvements.
• ✓Monitor trends in user behavior, harmful content types, and classifier drift to anticipate new risk patterns.