Staff Machine Learning Engineer, Home Surfaces

Spotify•Boston, MA

19d

About The Position

The Personalization team builds recommendation features that help Spotify listeners decide what to play next. The Surfaces Podcasts team owns podcast recommendations across prominent Spotify surfaces, including Home and Now Playing, helping listeners discover shows and episodes while helping creators reach the right audiences. As a Staff Machine Learning Engineer, you will shape personalized discovery and engagement at Spotify. You will work across recommendation systems, large language models, and production-scale machine learning infrastructure, taking models from research to production, setting technical direction in ambiguous problem spaces, and addressing personalization challenges at global scale.

Requirements

Deep experience building and deploying machine learning systems in production environments.
Deep experience with recommendation systems, ranking models, personalization, or large-scale content discovery platforms.
Proficiency in Python and hands-on experience building machine learning systems with PyTorch.
Experience with large language model training, fine-tuning, evaluation, and optimization, including SFT, distillation, and LoRA.
Experience with large-scale inference systems and the challenges of latency, reliability, and cost optimization.
Effective communication with technical and non-technical partners and ability to influence technical decisions beyond the immediate team.
Experience designing, executing, and interpreting online experiments and A/B tests to improve user outcomes.
Experience operating distributed machine learning workloads using technologies such as Ray, FSDP, HSDP, or similar frameworks.
Experience building and maintaining data pipelines and orchestration workflows using technologies such as Flyte, Airflow, BigQuery, and cloud-based storage platforms.

Responsibilities

Own and improve the machine learning models and systems that power the Home feed, including the Shortcuts experience.
Design, build, and ship personalized recommendations for Spotify listeners globally.
Build content recommendation systems for emerging agentic and AI-powered user experiences.
Train, fine-tune, evaluate, and optimize large language models using methods such as supervised fine-tuning, distillation, and parameter-efficient training.
Partner with Product Managers, Engineers, Data Scientists, and Designers to define and execute experimentation strategies.
Drive A/B testing, monitoring, model evaluation, and continuous optimization of recommendation quality, reliability, and cost efficiency.
Improve machine learning platform capabilities, data pipelines, and production systems supporting personalization at Spotify scale.
Drive technical direction in ambiguous problem spaces and contribute to the long-term architecture of personalization systems.
Mentor and support other Machine Learning Engineers, sharing knowledge and strengthening team engineering practices.