Machine Learning Engineer (Community Health) - TikTok BRIC Singapore

The Business Risk Integrated Control (BRIC) team is missioned to:
- Protect ByteDance users, including and beyond content consumers, creators, advertisers;
- Secure platform health and community experience authenticity;
- Build infrastructures, platforms and technologies, as well as to collaborate with many cross-functional teams and stakeholders.
The BRIC team works to minimize the damage of inauthentic behaviors on ByteDance platforms (e.g. TikTok, CapCut,Lark), covering multiple classical and novel community and business risk areas such as account integrity, engagement authenticity, anti spam, API abuse, growth fraud, live streaming security and financial safety (ads or e-commerce), etc.
Responsibilities
- Handling various business security issues on TikTok-related product lines (short video/live streaming/social engagement, etc.), and identifying fraudulent users, including those involved in pornography, fraud, gambling, and other violative activities.
- Utilize AI technologies such as CV/NLP/sequence/graph models to develop and deploy models across various TikTok scenarios, identifying risky users and content from massive amounts of images, videos, text, behaviors, and social relationships.
- Develop industry-leading risk control solutions to combat fraudulent users, analyze and mine potential data patterns, establish a perception system for attacks, and proactively identify business risks to safeguard the business.
Requirements:
Minimum Qualifications
- In-depth research in one of the fields of CV/NLP/Sequence/Graph, including but not limited to: image/video classification and recognition, object detection, unsupervised/semi-supervised learning, text classification, text mining, sequence modeling, graph models, etc.
- Familiar with one or more programming languages such as Python or C++, JAVA, with strong programming skills. Knowledge of big data processing technologies (Hive/Spark/Flink) and skilled at learning and applying industry-leading data architectures and technologies.
- Strong hands-on practical skills, familiar with the training and deployment of deep learning models, and able to apply algorithms to risk control business and verify their effectiveness.
- Excellent problem-analysis and problem-solving skills, with a passion for solving challenging problems.
Preferred Qualifications
- Prior experience in risk control algorithm research and development is preferred.