Machine Learning Engineer - Intern (TikTok BRIC) - 2027 Start (PhD)
Core
Build risk control algorithms and AI technologies to protect user safety, platform health, and ecosystem authenticity across TikTok and other ByteDance products.
Role type
PhD intern, research-focused machine learning engineer (risk control)
Builds
Risk control algorithms, AI models, and infrastructure for account and content integrity
Domain
Social media, e-commerce, advertising, and payment security
Deliverable
production ML models
Required skills
LLM/Agent systems engineering, Python, SQL, PyTorch/TensorFlow, graph learning, anomaly detection, multimodal models
Preferred skills
High-quality publications at top conferences, deep research in planning & reasoning, long task orchestration, RLHF, DPO, GRPO
Technologies
PyTorch, TensorFlow, LLMs, Agent tooling
Responsibilities
Build features and representation learning based on multi-source data to improve risk detection; Research and build the architecture of risk-control Agentic AI; Develop pre-trained models tailored for account security scenarios; Research long behavioral sequence models for risk identification.