The Personalization team makes deciding what to play next easier and more enjoyable for every listener. From Blend to Discover Weekly, we’re behind some of Spotify’s most-loved features. We built them by understanding the world of music and podcasts better than anyone else. Join us and you’ll keep millions of users listening by making great recommendations to each and every one of them.

We are looking for a Staff MLE to join Surfaces Podcasts. The Surfaces Podcasts team builds the systems that power podcast recommendations across some of Spotify’s most visible experiences, including Home and the Now Playing view. We work across candidate generation, ranking, and embedding models to help listeners discover their favorite new podcast and engage deeply with their favorite shows.

We’re also shaping the next generation of personalization through transformer-based models that bring more dynamic, context-aware recommendations to millions of listeners. You’ll collaborate closely with teams across Personalization, Experience, and the Podcast Mission to evolve podcast listening across Spotify.

What You'll Do

• Contribute to designing, scaling/building, evaluating, integrating, shipping, and refining reward signals for recommendations by hands-on ML development
• Promote and role-model best practices of ML systems development, testing, evaluation, etc., both inside the team as well as throughout the organization.
• Lead collaborations and align across PZN to integrate and A/B test mid-term signals in various recommendation systems

Who You Are

You have a strong background in machine learning, enjoy applying theory to develop real-world applications, with expertise in statistics and optimization, especially in sequential models, transformers, generative AI and large language models, and relevant fine-tuning processes.
You have hands-on experience with large cross-collaborative machine learning projects and managing stakeholders.
You have hands-on experience implementing production machine learning systems at scale in Java, Scala, Python, or similar languages. Experience with PyTorch, Ray, Hugging Face and related tools is required.
You have some experience with large scale, distributed data processing frameworks/tools like Apache Beam, Apache Spark, or even our open source API for it - Scio, and cloud platforms like GCP or AWS.
You care about agile software processes, data-driven development, reliability, and disciplined experimentation.

Where You'll Be

• We offer you the flexibility to work where you work best! For this role, you can be within North America as long as we have a work location.
• This team operates within the Eastern Standard time zone for collaboration