Description
The Personalization team at Spotify makes deciding what to play next on Spotify easier and more enjoyable for every listener. We seek to understand the world of music better than anyone else so that we can make great recommendations to every individual and keep the world listening.
The Surfaces Music team is responsible for music recommendations across Spotify's most visible surfaces, including Home and the Now Playing experience. We own music shelf and candidate generation as well as the ranking models that power these experiences. Our models include embedding models for deep catalog discovery, new release recommendations, and a unified transformer-based generative personalization model that is poised to reshape how we deliver personalized experiences across Spotify.
Responsibilities
- Contribute to the design, development, evaluation, and iteration of recommendation models , including candidate generation, ranking, and embedding models , powering music surfaces at scale.
- Drive hands-on ML development to improve reward signals and recommendation quality across Home, Now Playing, and other core surfaces.
- Contribute to the team's adoption of generative recommendation models, partnering with ML and AI infrastructure teams.
- Promote best practices in ML systems development, testing, and experimentation within the team.
- Collaborate with Data Science, Product, and Design partners to define success metrics, run A/B experiments, and translate insights into product improvements.
- Partner with teams across Personalization to integrate and test new signals in recommendation systems.
Requirements
- Strong background in machine learning and experience applying theory to real-world applications, with expertise in statistics and optimization , particularly sequential models, transformers, generative AI, and LLMs.
- Hands-on experience building and shipping production machine learning systems at scale, ideally in personalization or recommendation systems.
- Experience implementing ML systems in Java, Scala, Python, or similar languages. Familiarity with PyTorch, Ray or Hugging Face is a plus.
- Experience with large-scale distributed data processing frameworks such as Apache Beam, Apache Spark, or Scio, and cloud platforms like GCP or AWS.
- Experience collaborating across teams on complex ML projects and navigating cross-functional stakeholders.
- Care about agile software processes, data-driven development, reliability, and disciplined experimentation.
Benefits
- United States base range for this position is $210,000 - $260,000 plus equity.
- Health insurance, six month paid parental leave, 401(k) retirement plan, monthly meal allowance, 23 paid days off, 13 paid flexible holidays, paid sick leave.