Machine Learning Engineer II, Search Science and Data Infrastructure

Amazon Search creates powerful, customer-focused product search solutions and technologies. Whenever a customer visits an Amazon site worldwide and types in a query or browses through product categories, our systems go to work. We delight customers when we accurately understand their intent expressed via a query or image, and reflect that understanding throughout the search page - from layout to the search results and navigation. We make shopping effortless by helping customers easily explore our vast selection, narrowing down a myriad of options to a manageable consideration set while providing key information to make high confidence decisions with low post-purchase regret

Search Science Data Infrastructure team is responsible for delivering high quality and fresh ML model training data, and providing seamless access to all ML artifacts through federated Feature Store infrastructure. This big-data platform provides the ML training data to Amazon search ranking, matching quality, search economics and also powers live-site features, including search suggestions, query understanding, spelling, search result ranking, and personalization. Furthermore, 350+ teams across Amazon consume our datasets to power analytics and behavior models. We are located in downtown Palo Alto, a short walk from numerous shops and restaurants, and right across from the Caltrain station.


Key job responsibilities
As an ML Engineer you will:
- Lead development of services and infrastructure at the intersection of machine learning, big data, and distributed systems. Our products and services empower hundreds of science teams across Amazon to deliver machine learning at scale for ML model training, Feature engineering and Data quality monitoring.
- You will help manage machine learning lifecycle and operations using AWS AI services, DL compute resources, and our core search backend services for query understanding, semantic matching, and relevance ranking.
- You will build scalable data-intensive infrastructure that processes huge amounts of logs, catalogs, transactional data, and telemetry signals. By doing so, we enable teams to become more data-driven and build robust and explainable ML services.
- You will work with partners on data experimentation to advance Amazon product search, making it available across all geographic regions with variety of product search and discovery use cases across many categories.
- Lead the design, get your hands dirty and write code, and ultimately deploy big data and machine learning services. These services define the foundation of our search R&D processes, supporting science, product development and production of the world’s largest product search engine.
- Possess expert knowledge in performance, large scale distributed system scalability, system architecture, and engineering best practices.
- Obsess over operational excellence, evaluate system performance, security, design system metrics and driving quality improvements