Job Description

About the Team

Live-stream is a new and rapidly growing business that aims to bring joy to end users and allow more influencers to make an impact among their followers. And it's essential to pick the "right" live-stream for the "right" audiences. Our Live-stream Recommendation Infra team is responsible for building up and optimizing the infrastructure for such recommendation systems, so as to provide the most stable and best experience for our users. We work closely with applied machine learning engineers and build scalable systems to support all kinds of innovative algorithms and techniques.

Responsibilities

- Design and build backend systems that support large-scale recommendation workloads, including training, inference, and data pipelines.

- Develop robust and efficient model infrastructure, including distributed training pipelines and low-latency inference serving.

- Architect and improve data pipelines to enable efficient collection, preprocessing, and offline feature engineering for recommendation and ranking models.

- Collaborate closely with ML engineers and researchers to productionize models and integrate them into the TikTok Live recommendation stack.

- Drive performance optimization and cost-efficiency across training, inference, and data workflows.

- Ensure system robustness, scalability, and maintainability in high-traffic live streaming scenarios.