Staff Machine Learning Engineer, Home Surfaces
Core
Building and improving recommendation systems for podcast discovery on Spotify's Home and Now Playing surfaces using large language models and production-scale ML infrastructure.
Role type
Staff Machine Learning Engineer (Recommendation Systems & LLMs)
Builds
Personalized podcast recommendations and AI-powered user experiences for millions of global listeners
Domain
Music streaming, Podcasts, Large Language Models, Recommendation Systems
Deliverable
production ML models
Required skills
Recommendation systems, Large Language Models (LLMs), Python, PyTorch, Supervised Fine-Tuning (SFT), Model Distillation, LoRA, Distributed ML workloads (Ray, FSDP, HSDP), Data pipelines (Flyte, Airflow, BigQuery), A/B testing, Technical leadership
Preferred skills
None stated
Technologies
PyTorch, Ray, FSDP, HSDP, Flyte, Airflow, BigQuery
Responsibilities
Own and improve ML models powering the Home feed and Shortcuts experience; Design and ship personalized recommendations; Build content recommendation systems for agentic/AI experiences; Train and optimize LLMs; Partner with cross-functional teams on experimentation; Drive A/B testing and model optimization; Improve ML platform capabilities; Drive technical direction and architecture; Mentor other ML Engineers
Seniority
Staff, hands-on IC with mentorship
