Senior Machine Learning Engineer II, Search Retrieval

What’s the role?

Etsy is hiring a Senior Machine Learning Engineer II in the Search Retrieval team. We develop efficient, state-of-the-art, large-scale search models that sift through a universe of over a hundred million special, extraordinary items to help Etsy buyers find something they will love.

In Search Retrieval, you'll closely follow the latest research and implement its best ideas efficiently, at scale for the benefit of Etsy's buyers. Working hands-on with a cross-disciplinary team, you will deliver robust production models that will be A/B tested live on Etsy.com! Does helping people find the perfect handmade or vintage good from using the latest ML research sound exciting to you? If so, we could be the perfect match.

This is a full-time position reporting to the Senior Engineering Manager for the Search Retrieval team and the base salary range will be 184,000 - 216,000 USD per year. In addition to salary, you will also be eligible for an equity package, an annual performance bonus, and our competitive benefits that support you and your family as part of your total rewards package at Etsy. For this role, we are considering candidates based in the U.S. who are either remote, flex, or office-based. Etsy offers different work modes to meet the variety of needs and preferences of our team. Learn more about our Flex and Office-based work modes and workplace safety policies here.

What’s this team like at Etsy?

We Apply The Latest Techniques From Machine Learning, Natural Language Processing, And Information Retrieval Research At Etsy’s Scale, Including

Search Retrieval is a product engineering team that strives to improve the relevancy of search results for Etsy buyers.

• Distributed, highly efficient random walk based graph recommendation engine written in Rust.
• Graph neural networks that learn from billions of previous interactions between our buyers and sellers.
• Inverted indexes that incorporate the latest research in sparse retrieval.
• Deep neural dense retrieval models powered by Transformers and fast approximate nearest neighbor (ANN) search

What does the day-to-day look like?

In Our Team, You Will

Our team of applied scientists, ML engineers, data analysts, and product managers work as a cross-disciplinary team in a primarily remote setting.

• Strategically pick ideas from the latest in ML research with the highest potential to benefit Etsy buyers and quickly prototype them on Etsy data.
• Propose pragmatic long-term project plans that demonstrate value early through fast iteration.
• Create large, scalable data ETL pipelines for new features to power our models.
• Optimize ML models to meet latency SLAs at the scale of Etsy production traffic and launch live A/B experiments to test our hypotheses.
• Analyze and learn from experiment results, and disseminate our findings throughout Etsy and beyond through conference publications and talks.
• Lead high code quality and engineering standards in the team.
• Of course, this is just a sample of the kinds of work this role will require! You should assume that your role will encompass other tasks, too, and that your job duties and responsibilities may change from time to time at Etsy's discretion, or otherwise applicable with local law.

Qualifications

Qualities that will help you thrive in this role are:

It Also Helps If You Have

You are a caring teammate who enjoys helping others. Your growth mindset means you love feedback and find joy in continuous learning and self-improvement. You care about how ML impacts real people.

• A track record of efficiently implementing ML models for high traffic, low latency, large data applications that produced meaningful impact on the end users.
• 2+ years experience writing high quality production code, demonstrating strong software engineering and coding skills.
• Proficiency in neural ML frameworks such as PyTorch, Jax, Keras, Tensorflow.
• Experience with large-scale data ETL frameworks and ML workflow frameworks such as Apache Spark, Beam, Airflow, Kubeflow Pipelines etc.
• Proficiency in Python, Scala, and/or other equivalent languages.
• Experience in Rust (our graph engine is written in Rust!), Golang, C/C++ or other high-performance programming languages.
• A product focus and passion for using machine learning to tackle real-world problems.
• Effective written communication and comfort working in a remote team with a strong Slack and Google Docs culture.
• Research experience and publications in one of the following fields: information retrieval, recommendation systems, knowledge graphs, natural language processing.
• Experience in implementing Graph Neural Networks (GNNs) in a large-scale search and/or recommendation systems context