What You'll Be Doing

• Define risk categories and severity levels (e.g., scams/phishing, traffic diversion, investment misinformation, harassment/hate, sexual/violent content, IP infringement).
• Build multi-language lexicons and rule strategies (thresholds, combinations, rate limits) for key scenarios: posts, comments, group chats, livestreams, profiles.
• Integrate risk signals and systems: account risk score, device/behavior signals, reports, human review, content detection models.
• Design enforcement and visibility actions: downrank/limit reach, self-only/limited visibility, global removal, interaction limits, account penalties.
• Design the flagged-user experience: clear notices, verification steps when needed, appeals flow, result notifications, SLAs.
• Create dashboards and KPIs (violation rate, false positives, appeal pass rate, repeat-offense rate, handling time, review cost) and continuously optimize.