Senior Product Manager, AI Safety Evaluation & Governance - TikTok Safety Product

What you'll do

• Define model evaluation frameworks, quality gates, regression thresholds, and monitoring requirements.
• Build severity-stratified evaluation datasets and benchmarks; analyze model results and improve LLM-based evaluation workflows.
• Align Algorithm, Policy, Recommendation, Labeling, and Data Science teams on improvements and scale evaluation operations.

What they ask for

• Bachelor’s degree or above in Statistics, Computer Science, Data Science, or a closely related quantitative field.
• 5+ years in AI/ML evaluation, model quality assurance, trust & safety product, or data-driven product management, including designing evaluation systems or ground-truth datasets.
• Proficiency in SQL for statistical analysis, data pipeline work, and ad-hoc investigation; professional experience with LLMs, prompt engineering, LLM-based evaluation, and iterative optimization of agentic workflows.
• Deep content safety or trust & safety knowledge, cross-functional leadership, and written and spoken Chinese fluency.

Nice to have

• Experience building evaluation frameworks for large-scale content moderation or safety AI systems.
• Familiarity with adversarial content, annotation quality systems, recall-oriented evaluation, recommendation or ranking, large-scale labeling, or agentic AI workflows.

Tools

SQL, LLMs

AI focus

Evaluates and improves AI-powered content safety models, including LLM-based evaluation, prompt engineering, LLM-assisted labeling, and agentic workflows.