Senior Machine Learning Engineer - Policy & Safety

Spotify•New York, NY

2d•Hybrid

About The Position

The Policy & Safety team sits within Content Platform in the Experience Mission, building the systems that keep Spotify safe, compliant, and trusted by millions of users and creators. This team owns Spotify’s content moderation infrastructure — from detection models to policy enforcement systems and compliance data pipelines. Working at the intersection of machine learning, platform engineering, and regulatory compliance, the team partners closely with Trust & Safety, Legal, and Public Affairs. They’re on the critical path for every new content type and social feature — including messaging, comments, and collaborative experiences — ensuring safety is built in from day one. With a strong focus on “safety by default,” the team is investing in large-scale rearchitecture and ML-driven systems to proactively protect users and empower safer interactions across the platform.

Requirements

Solid experience building and deploying machine learning systems in production environments at scale
Experienced with training, evaluating, and maintaining ML models using modern frameworks such as PyTorch
Deep understanding of machine learning evaluation, including dataset design, metrics, and continuous improvement systems
Knowledge of how to design systems that balance performance, reliability, and real-world impact in high-stakes domains
Experience leading technical projects and influencing direction within a team or product area
Experience with distributed systems or backend technologies (e.g., Scala)

Nice To Haves

Care about building safe, responsible, and user-centric ML systems
Comfortable working across disciplines, partnering with legal, policy, and product stakeholders

Responsibilities

Design, build, and ship production-grade machine learning systems that power content safety and policy enforcement at Spotify scale
Own and lead key technical initiatives across detection, classification, and policy evaluation systems
Develop and maintain ML models for content moderation, including multimodal and LLM-based systems
Build robust evaluation frameworks, including standardized datasets, offline and online metrics, and continuous improvement loops
Drive experimentation to improve model performance, reliability, and fairness in safety-critical systems
Collaborate closely with cross-functional partners in Trust & Safety, Legal, and Public Affairs to align on policy and enforcement needs
Provide technical leadership within the team, mentoring engineers and contributing to ML strategy and prioritization
Represent technical decisions and trade-offs in stakeholder discussions and influence product direction