Twitch uses machine learning to personalize which live streams appear on each viewer's homepage, recommended channels sidebar, and mobile Discovery Feed, processing billions of requests daily to match viewers with streams based on viewing history and engagement patterns.

Twitch's "State of Engineering 2023" blog details the ML infrastructure: a Machine Learning Feature Store (MLFS) for centralized feature sharing, Real-Time Feature tools built on AWS Kinesis with seconds of latency, a Model Registry for model versioning, and an automated Model CI/CD pipeline with canary testing and rollback. In July 2024, Twitch launched a redesigned mobile app with a vertical-scroll Discovery Feed as the default landing experience. Tags (introduced September 2018) serve as inputs to the recommendation engine. witch's 2021 Research Fellowship listed "recommender systems" including "preference elicitation, algorithmic scalability, context-aware recommendations, user modeling, algorithmic bias, and ethics of recommender systems" as core research areas.