内容安全算法工程师-TikTok业务风控
Core
Build end-to-end risk control systems and intelligent algorithms to protect TikTok users, creators, and business clients from content and financial risks in live streaming and other scenarios.
Role type
Senior IC machine-learning engineer (risk control & content safety)
Builds
High-precision, low-latency risk identification models for TikTok's live streaming ecosystem
Domain
Social media platform risk control, content safety, financial fraud prevention
Deliverable
production ML models
Required skills
Machine learning theory, deep learning, classification/clustering/sequence prediction algorithms (XGBoost, DeepFM, Transformer), Python/C++/Go, TensorFlow/Pytorch, distributed system development
Preferred skills
Agentic AI (LLM agents), LLM post-training (RLHF, SFT), multimodal representation learning, video understanding, cross-modal retrieval, internet community risk control experience
Technologies
XGBoost, DeepFM, Transformer, TensorFlow, PyTorch, LLMs
Responsibilities
Design algorithm systems for ecosystem safety and financial risk control in live streaming; Build and optimize risk identification models using multimodal data; Analyze black-market attack patterns and design robust countermeasures; Apply Agentic AI and multimodal pre-training for automated decision-making
Seniority
Senior, hands-on IC
