Machine Learning Engineer II, Fulfillment

etsy; United States; On-site; USD 120,000 - 190,000; Full time

Job summary

Etsy is hiring a Machine Learning Engineer II to join the Fulfillment ML team. You will work on models powering Etsy's shipping and delivery estimates, pricing, and fulfillment decisions at scale, collaborating with engineers, data scientists, and product teams.

The role combines ML platform engineering—building pipelines and tooling—with applied ML work, including model evaluation, backtests, tuning, and readiness reviews.

Qualifications

• Experience building ML pipelines and deploying models to production.
• Familiarity with end-to-end ML lifecycle: training, validation, promotion, inference.
• Collaborative mindset with cross-functional partners.

Responsibilities

• Build and maintain ML training, validation, promotion, and batch‑inference pipelines.
• Extend Airflow DAGs for model workflows across environments.
• Collaborate on backtests, tuning, and readiness reviews with team.
• Assist migrations to internal ML platform and config wiring.
• Debug production incidents affecting model-driven surfaces.
• Improve observability and data quality tagging to catch issues early.

Skills

Python

PyTorch

Airflow

Spark

Ray

BigQuery

GCS

Vertex AI

Dataproc

Tools

Python

PyTorch

Airflow

Spark

Ray

BigQuery

Google Cloud

Vertex AI

Dataproc

Job description

Company Description

Etsy is the global marketplace for unique and creative goods. We build, power, and evolve the tools and technologies that connect millions of entrepreneurs with millions of buyers around the world. As an Etsy employee, you will tackle unique, meaningful, and large-scale problems alongside passionate coworkers, all the while making a rewarding impact and Keeping Commerce Human.

We believe exceptional companies are built by exceptional teams, and we're intentional about it. That means hiring great people, setting them up for success from day one, and giving them real reasons to grow their careers here. We invest in development that goes beyond promotions, and we foster the trust and relationships that help people do their best work together.

What's the role?

Etsy is hiring a Machine Learning Engineer II to join the Fulfillment ML team. Our team owns the machine learning that powers Etsy's shipping and delivery experience - models that help buyers see accurate delivery estimates, help sellers price and ship their orders accurately, and help Etsy make smart fulfillment decisions at scale. Examples of models we own include Estimated Delivery Date (EDD) prediction, shipping price prediction, transit-time modeling, and delivery-risk signals. We're a multi-functional group of engineers, applied scientists, and product managers, and we partner closely with data scientists, designers, and product-facing engineering teams across Etsy.

This role combines ML platform engineering with hands-on applied ML work. On the platform side, you'll build and operate the pipelines, orchestration, and tooling that take our models from training through validation, promotion, and inference in production. On the applied side, you'll partner with senior teammates on model evaluation, backtests, tuning, and readiness reviews - you're not expected to design new model architectures yourself, but you will work deeply with the models, learn how they behave, and grow your applied ML expertise over time.

This is a great opportunity for a strong software engineer early in their ML career who wants to level up on both the systems side and the modeling side, inside a mature MLOps stack that ships model predictions to millions of Etsy buyers and sellers every day.

This is a full-time position reporting to the Engineering Manager, Fulfillment ML.

What's this team like at Etsy?

Fulfillment ML owns the machine learning behind Etsy's shipping and delivery experience - across buyer-facing surfaces like Search, Listing, Cart, and Checkout, and across seller-facing tools. Our team includes MLEs, applied scientists, and platform engineers. New model design is usually led by senior teammates; the surrounding applied ML work (evaluation, backtests, tuning, readiness) and the platform/infrastructure work are shared responsibilities across the team, including this role. Our stack includes Python, PyTorch, Ray, Spark, Airflow, Google Cloud's Gemini Enterprise Agent Platform (formerly Vertex AI), Dataproc, BigQuery, and GCS.

Here's a taste of the problems we're solving:

• How do we predict when an order will arrive at a buyer's door across dozens of countries, carriers, and shipping methods?
• How do we tell buyers what's likely to arrive by a given date without underpromising or overpromising?
• How do we spot delivery risk early enough to help sellers and buyers make good decisions?
• How do we retrain, validate, and promote models weekly without breaking anything downstream?
• How do we notice a model is drifting before buyers or sellers do?
• How do we roll back a bad model in minutes, not hours, with confidence?
• How do we predict shipping costs so buyers know what to expect at checkout and sellers can plan fulfillment with confidence?
• How do we make our shipping and fulfillment signals feel consistent everywhere they show up in the buyer journey?

What does the day-to-day look like?

• Build and maintain the training, validation, promotion, and batch-inference pipelines that put our models into production and keep them healthy there.
• Write and extend Airflow DAGs within our templatized MLOps framework - training, inference, and validation workflows across staging, shadow, and production environments.
• Partner with teammates on the applied ML side - running backtests, tuning model outputs to hit business targets (on-time delivery rate, cost, accuracy), and supporting seasonal readiness reviews.
• Contribute to model migrations onto our internal ML platform - wiring up model configs, feature pipelines, postprocessors, and inference runners alongside the teammates who design the models.
• Triage and debug production incidents that affect model-driven surfaces (for example, delivery estimates on Cart/Checkout or shipping prices in Search results), and coordinate fixes with partner teams.
• Improve our observability - dashboards, retrain trending, data-quality tags, and alert hygiene - so problems are easier to spot before they reach buyers or s