Machine Learning Engineer II, Fulfillment

Overview You join Etsy’s Fulfillment ML team to advance machine learning for shipping and delivery. You’ll work on production ML pipelines and collaborate with engineers, data scientists, and product partners to improve delivery estimates, pricing, and fulfillment decisions at scale. The role blends platform engineering with applied ML, offering growth in an established MLOps stack. You’ll contribute to model health, backtests, and readiness while building robust, observable systems that serve buyers and sellers worldwide. This is a chance to impact how orders arrive on time across many markets. Compensation / Benefitsequity packageannual performance bonuscompetitive benefitssalary based on location and experienceremote/hybrid optionsgeographic pay zones Responsibilities Build and maintain training, validation, promotion, and batch-inference pipelines for production models Write and extend Airflow DAGs within a templated MLOps framework across staging, shadow, and production Partner with applied ML teammates on backtests, tuning, and readiness reviews to hit business targets Contribute to model migrations onto the internal ML platform (configs, feature pipelines, postprocessors, inference runners)Triage and debug production incidents affecting model-driven surfaces and coordinate fixes Improve observability with dashboards, retrain trends, data-quality tags, and alert hygiene Write clear PRs, design docs, and runbooks to durableize knowledge Participate in on-call rotation for model-health and data-quality alerts Assume broader responsibilities as needed and adapt to changing priorities Key requirements 1–3 years of professional experience shipping production ML systems Strong Python skills and good code hygiene Exposure to batch data pipelines and ML orchestration (Airflow, Kubeflow, Dagster, Prefect)SQL and at least one cloud data warehouse (Big Query, Snowflake, Redshift)Understanding of ML lifecycle: data, metrics, validation, production deployment Clear, testable, maintainable code with openness to feedbackcollaboration across multi-functional teamseffective communicationeagerness to learn and be mentored PyTorch or another deep learning framework (bonus)distributed compute exposure (Spark, Ray, Dask) (bonus)experience with orchestration or feature pipelines (bonus)