Designing Feed Selection in Twitter-Style Systems

Ever wondered how social media platforms decide exactly which posts appear on your timeline? Behind every scroll is a complex system design that filters billions of candidates down to a handful of highly engaging updates. This text-only course guides you through the foundational architecture of feed recommendation systems, using the mechanics of Twitter's feed system as a blueprint. You will understand how algorithms balance content from people you follow with popular global posts to keep timelines fresh and relevant. What you'll learn: - Understand the core stages of feed generation, from candidate generation to final heavy ranking. - Analyze how systems select in-network content from users you follow and out-of-network content from the wider platform. - Explore the role of machine learning models in predicting user engagement and ranking posts. - Learn how filtering mechanisms handle blocked accounts, muted keywords, and duplicate content. - Discover modern system design patterns, including vector search and real-time graph processing. The course begins with foundational definitions of feed architectures before breaking down the step-by-step pipeline of content retrieval, filtering, and scoring. You will consolidate your knowledge through written conceptual exercises and architectural analysis. This course is designed for beginner software engineers, system design aspirants, and curious tech enthusiasts, with no prior experience with recommendation algorithms required. Start reading today to demystify the algorithms shaping our digital experiences.