Senior Product Manager - AI Safety & Governance

TikTok; San Jose (CA); On-site; USD 186,000 - 374,000; Full time

Job summary

TikTok is seeking a Senior Product Manager, AI Safety Evaluation & Governance, to lead the design and governance of AI safety evaluation frameworks. You will work with Safety, Algorithm, and Policy teams to define recall-based metrics, create diverse ground-truth datasets, and steer model improvement roadmaps across languages and regions.

The role focuses on evaluating AI systems for content safety at scale, building robust pipelines for rigorous testing, and ensuring alignment with policy

Qualifications

• Bachelor's degree or above in a quantitative field (Statistics/CS/Data Science)
• 5+ years in AI/ML evaluation or trust-safety product
• Strong data analysis skills and SQL proficiency
• Proven experience with LLMs, prompting and evaluation workflows
• Deep understanding of safety/content moderation and risk assessment
• Strong cross-functional leadership without direct authority
• Fluency in written and spoken Chinese for China market

Responsibilities

• Define and own the evaluation methodology for AI-powered safety models with measurable quality signals.
• Design diverse evaluation datasets and sampling strategies to cover real-world content.
• Create severity-ranked benchmarks to assess model performance across risk levels.
• Conduct hands-on data analysis and model evaluations, reporting actionable insights.
• Develop prompts and LLM-based evaluation pipelines, and improve agentic workflows.
• Coordinate with policy, algorithm, labeling, and data science teams to align roadmaps.

Skills

AI Safety

Evaluation

SQL

LLMs

Cross-functional Leadership

Data Analysis

Education

Bachelor's degree in Statistics, Computer Science, Data Science

Tools

LLM-based evaluation

Ground truth labeling tools

Job description

TikTok is seeking a Senior Product Manager, AI Safety Evaluation & Governance, to lead the design and governance of AI safety evaluation frameworks. You will work with Safety, Algorithm, and Policy teams to define recall-based metrics, create diverse ground-truth datasets, and steer model improvement roadmaps across languages and regions.

The role focuses on evaluating AI systems for content safety at scale, building robust pipelines for rigorous testing, and ensuring alignment with policy