We’re hiring Machine Learning Engineers at the Senior and Staff levels to join one of our Search and Discovery teams: Search Ads, Recommendations Ranking, or Search Retrieval. You will build user-centric ML systems for a two-sided marketplace with 90+ million buyers, 6+ million sellers, and hundreds of millions of unique, handmade, and vintage listings. We’re looking for people who are product- and delivery-driven and passionate about applying ML innovations (ranking, retrieval, deep learning, reinforcement learning, LLMs, and more) to improve the Etsy
buyer and seller experience.
You’ll own models end-to-end, from data and feature pipelines through training, evaluation, and production serving, and validate their impact through live A/B experiments on Etsy.com and the Etsy app. Team placement is based on your background, level, and interests, and will be discussed during the interview process.
This is a full-time position reporting to the Engineering Manager of the team you join.
What’s this team like at Etsy?
We apply cutting-edge machine learning, deep learning, and information retrieval techniques to high-impact problems at Etsy scale. Depending on your experience, you may be considered for one of the following teams:
●Search Ads: Builds the ML systems behind our sponsored search product, including retrieval, ranking, ad allocation, and user engagement modeling. We ensure the right ads reach the right buyers, personalize the ad experience to each buyer’s journey, and help sellers get the most from their advertising budgets, all on a pipeline that scales to our 90+ million buyers and sellers.
●Recommendations Ranking: Develops ML models that understand buyers, personalize, and rank recommendations across Etsy.com and the Etsy app. Work includes user modeling, personalized ranking, multi-objective optimization, intent and lifecycle prediction, and keeping feeds fresh and reactive to user actions, emerging trends, and seasonal shifts.
●Search Retrieval: Builds efficient, scalable, and intelligent retrieval systems that surface the most relevant listings from a catalog of over a hundred million unique items. We combine lexical matching, semantic retrieval, and hybrid models to power the first step inEtsy’s search stack, using techniques such as dual-encoder models, vector similarity
search, and query expansion.