LinkedIn Deploys AI Generative to Boost Feed Relevance

LinkedIn is rolling out advanced AI-powered generative recommendation systems to deliver more relevant content across its platform. The update, detailed by Chief Technology Officer Erran Berger, marks a significant evolution in how the professional network understands and responds to user behavior.

Previously, LinkedIn relied on separate ranking models for different parts of the platform, such as the main feed, job recommendations, and advertising. The new approach uses large-scale generative recommenders and sequence models to view user activity as one continuous professional journey rather than isolated interactions.

This unified system allows signals from one area of the platform to inform recommendations elsewhere. For example, engagement with content in the feed can now influence job suggestions, notifications, and even connection recommendations. Berger noted that professional identity develops over time, and the new models better reflect that reality.

“With advances in generative recommenders and large-scale sequence models, we’re rethinking the foundation of how recommendations work,” Berger explained.

The scale of the challenge is substantial. LinkedIn processes more than 1.8 million feed updates viewed per minute. AI systems enable the platform to evaluate far more candidate posts quickly and surface more relevant opportunities to users.

The shift is designed to increase overall engagement by showing content that aligns better with each member’s broader professional goals and interests. By expanding the pool of potential content and connecting activity across the entire platform, LinkedIn aims to create a more cohesive and useful experience.