Pinterest Replaces Journey Clustering with 4B Qwen3 LLM, Reports Improved Engagement

InfoQ (AI, ML & Data)
Read full story· 3 min
Pinterest replaced its multi-stage clustering system for user journey inference with a fine-tuned 4-billion-parameter Qwen3 language model, improving email click-through by 1.1% and push notification opens by 1.3%. The model analyzes up to 150 user actions over a year to identify coherent long-term goals, enhancing recommendation relevance.

More in Business & Enterprise