Responsibilities

Business Risk Integrated Control (BRIC) is a global team protecting ByteDance users, platform health, and the authenticity of community and business ecosystems. We build the risk control algorithms, AI technologies, platforms, and infrastructure behind that work, across products including TikTok, CapCut, Lemon8, and Lark. Our work spans account and content integrity, and safety across advertising and e-commerce. We apply machine learning across areas such as graph learning, anomaly detection, and multimodal models to understand users, content, behavior, and the relationships between them. The challenge is recognizing genuine activity accurately, so the people using our products never notice the systems working. You will shape the trust, safety, and sustainable growth of ByteDance's global platforms, applying advanced AI to some of the hardest real-world problems in the field. TikTok's Emerging Businesses encompass fast-growing products and innovative business models, including TikTok Drama, TikTok AI, and TikTok Open Platform, spanning content creation, AI capabilities, open ecosystems, and monetization opportunities. Operating in complex global regulatory and highly adversarial environments, the team focuses on critical risks such as AI feature and model capability abuse, advertising fraud and fake conversions, traffic and engagement manipulation, large-scale account farming, and account abuse. Our mission is to build scalable and intelligent risk control systems across AI safety, advertising anti-fraud, traffic integrity, account security, and platform capability protection, safeguarding sustainable business growth, maintaining a trustworthy commercial ecosystem, and ensuring the secure and compliant use of platform resources. Responsibilities - End-to-end Risk System Development: Lead the design and development of risk control algorithms and systems across TikTok Drama, TikTok AI, TikTok Open Platform, and other emerging businesses, focusing on detecting and mitigating AI capability abuse, advertising fraud, fake conversions, traffic and engagement manipulation, account farming, account hijacking, and API abuse. - Multimodal Risk Model Development & Optimization: Develop high-precision, low-latency, and reliable real-time risk detection and scoring models using large-scale multimodal data, including account attributes, device fingerprints, content signals, AI feature usage logs, advertising conversion funnels, user behavior sequences, and API invocation logs. - Adversarial Defense & Risk Discovery: Analyze sophisticated abuse patterns, identify emerging threats through data analysis and risk mining, and develop robust adversarial strategies and hybrid rule-model defense systems to protect emerging businesses against evolving attacks. - Cutting-edge Research & Implementation: As a technical owner, drive technology research, solution evaluation, architecture design, proof-of-concept (PoC) validation, and production implementation. Explore and apply advanced technologies, including Agentic AI, LLM post-training and fine-tuning, multimodal representation learning, behavioral sequence modeling, and graph learning, to enable automated risk discovery, investigation, reasoning, and decision-making. - Cross-functional Collaboration: Own the full lifecycle of algorithm development, from technical architecture and model evaluation to deployment and continuous optimization. Collaborate with engineering, product, advertising, growth, strategy, moderation, and data teams to drive technical innovation and establish scalable, reusable risk control capabilities across emerging business scenarios.

Qualifications

Minimum Qualification(s) - Master's degree or higher in Computer Science, Mathematics, Statistics, Artificial Intelligence, or a related field. - Algorithmic Foundation: Strong theoretical foundation in machine learning and deep learning, with proficiency in classification, ranking, sequence modeling, and anomaly detection techniques, including XGBoost, DeepFM, Transformers, and Graph Neural Networks. - Engineering Skills: Strong programming skills in at least one of Python, C++, or Go, with hands-on experience in PyTorch or TensorFlow. - Core Competencies: Strong analytical thinking, data intuition, and a passion for adversarial risk control and complex problem-solving. Demonstrated ability to design technical solutions, validate proofs of concept (PoCs), drive technical initiatives, and collaborate effectively across teams. Preferred Qualification(s) - PhD in Computer Science, Artificial Intelligence, Machine Learning, Mathematics, Statistics, or a related field is preferred, particularly with research experience in machine learning, deep learning, or AI-related domains. - Agentic AI Experience: Experience building LLM agents, integrating tools, planning complex tasks, and deploying automated risk investigation and decision-making solutions. - Multimodal & Behavioral Modeling: Experience in content understanding, behavioral sequence modeling, device fingerprinting, account anomaly detection, advertising fraud detection, or traffic manipulation detection. - Industry Experience: Proven experience developing machine learning algorithms for AI safety, advertising anti-fraud, traffic integrity, account risk control, Open Platform API risk control, or internet ecosystem integrity, with experience developing large-scale distributed risk control systems, offline training pipelines, and online inference systems. - End-to-end experience in risk technology research, proof-of-concept (PoC) validation, and scaling solutions from initial pilots to large-scale production environments.

Job Information