When users ask for “music for a solo night drive through the city,” recommender systems face a fundamental challenge: popular defaults often fail to capture the nuance of situational queries. At Spotify, our goal is to build systems that learn listening preferences directly from every play, skip, save, and refinement, treating each interaction as preference feedback. This loop is especially important for subtle requests: if a user skips a high-energy pop track but listens through a moody electronic one, that behavior should become a clear signal of the desired vibe. Thus, the key question is: how do we build such a system that can create great recommended music playlists from user requests and which is capable of continuously learning from feedback to deliver better recommendations over time? In this post, we present our approach using LLM-based agentic systems that interpret queries, orchestrate tools, and adapt through preference learning, going beyond traditional recommender systems to enable more nuanced, intent-driven playlist generation.

Spotify_Research's publication is a hub for academic research and industry insights in the field of music streaming technology and user behavior analysis. Through research papers, data analysis, and experimental studies, Spotify_Research offers  insights into music recommendation algorithms, user engagement metrics, and content discovery mechanisms. Readers can learn about the latest advancements in music streaming technology, personalized playlist generation, and user-centric design principles to enhance the music listening experience. Additionally, Spotify_Research provides data-driven insights, user studies, and market analysis to help music enthusiasts and industry professionals understand the evolving landscape of digital music consumption and anticipate future trends.

Spotify Research

Spotify Research presents a hybrid approach for personalizing AI-powered music recommendations using LLM-based agentic systems. The method combines reward models with Direct Preference Optimization (DPO) to create a continuous learning flywheel that adapts to user preferences from listening behavior. The system interprets natural language queries, orchestrates music search tools, and learns from user interactions like plays, skips, and saves. Production A/B tests showed 4% increase in listening time, higher playlist saves, and 70% reduction in erroneous tool calls while maintaining quality standards.

Personalizing Agentic AI to Users' Musical Tastes with Scalable Preference Optimization

A hybrid approach: Reward Models + Direct Preference Optimization

Engineering practices that made the difference