About this role
About the team
We are the Data Scientists dedicated to Creator Governance for TikTok Shop. We act as the strategists and health monitors of the ecosystem. We combine data analysis, lightweight machine learning, and metric construction to solve critical business problems. We define what "Healthy" means for the platform, build the strategies to enforce it, and visualize the pulse of our global creator community.
Why this role?
- Hybrid Impact: You won't just analyze data; you will build strategies (rules/models) to stop bad actors and design the metrics that guide the entire team.
- Full Cycle Ownership: From discovering a new risk pattern -> building a detection strategy -> monitoring its success via dashboards.
- Ecosystem Guardian: You will be the first to know when the ecosystem health fluctuates and the one who identifies why.
What You’ll Do
1. Governance Strategy & Risk Mining
- Pattern Discovery: Deep dive into creator behaviors and content logs to identify new abuse patterns (e.g., automated spam, matrix accounts, engagement farming).
- Strategy Construction: Translate insights into executable governance strategies using Rules, Thresholds, and Risk Scores. Quickly deploy tactics to mitigate sudden attacks.
- Practical Machine Learning: Develop agile, offline models (e.g., XGBoost, Random Forest, Anomaly Detection) and use Clustering to uncover organized bad-actor syndicates that simple rules might miss.
2. Metric System & Dashboard Construction
- North Star Definitions: Define and standardize core metrics for Creator Governance, such as "Low-Quality Content Rate," "Creator Violation Rate," and "Ecosystem Health Index."
- Dashboarding: Build and maintain comprehensive data dashboards (using Tableau, Metabase, or internal tools) to visualize risk trends, strategy performance, and daily operational status.
- Automated Monitoring: Design alert systems that automatically flag anomalies in the data (e.g., a sudden spike in new account registrations or a drop in high-quality content supply).
3. Performance Tracking & Root Cause Analysis
- Strategy Evaluation: Continuously monitor the Precision and Recall of governance strategies. Analyze false positives to minimize impact on good creators.
- Deep-Dive Analysis: When core metrics fluctuate (e.g., AIGC impact, GMV drops or complaint rates rise), conduct rapid root-cause analysis to determine if it is caused by governance interventions or external factors.
We are the Data Scientists dedicated to Creator Governance for TikTok Shop. We act as the strategists and health monitors of the ecosystem. We combine data analysis, lightweight machine learning, and metric construction to solve critical business problems. We define what "Healthy" means for the platform, build the strategies to enforce it, and visualize the pulse of our global creator community.
Why this role?
- Hybrid Impact: You won't just analyze data; you will build strategies (rules/models) to stop bad actors and design the metrics that guide the entire team.
- Full Cycle Ownership: From discovering a new risk pattern -> building a detection strategy -> monitoring its success via dashboards.
- Ecosystem Guardian: You will be the first to know when the ecosystem health fluctuates and the one who identifies why.
What You’ll Do
1. Governance Strategy & Risk Mining
- Pattern Discovery: Deep dive into creator behaviors and content logs to identify new abuse patterns (e.g., automated spam, matrix accounts, engagement farming).
- Strategy Construction: Translate insights into executable governance strategies using Rules, Thresholds, and Risk Scores. Quickly deploy tactics to mitigate sudden attacks.
- Practical Machine Learning: Develop agile, offline models (e.g., XGBoost, Random Forest, Anomaly Detection) and use Clustering to uncover organized bad-actor syndicates that simple rules might miss.
2. Metric System & Dashboard Construction
- North Star Definitions: Define and standardize core metrics for Creator Governance, such as "Low-Quality Content Rate," "Creator Violation Rate," and "Ecosystem Health Index."
- Dashboarding: Build and maintain comprehensive data dashboards (using Tableau, Metabase, or internal tools) to visualize risk trends, strategy performance, and daily operational status.
- Automated Monitoring: Design alert systems that automatically flag anomalies in the data (e.g., a sudden spike in new account registrations or a drop in high-quality content supply).
3. Performance Tracking & Root Cause Analysis
- Strategy Evaluation: Continuously monitor the Precision and Recall of governance strategies. Analyze false positives to minimize impact on good creators.
- Deep-Dive Analysis: When core metrics fluctuate (e.g., AIGC impact, GMV drops or complaint rates rise), conduct rapid root-cause analysis to determine if it is caused by governance interventions or external factors.