Our mission is to reinvent how customers discover products and services, and shop.
We want customers to think of Alexa as their personal shopping assistant, providing a better experience than in any shop or store, saving our customers’ time, and help them make better shopping decisions. We need to respond to any and all customer questions with a useful answer, no matter how personal or subjective. For example:
"Alexa... are Fitbits waterproof? ...what is the most popular HDTV on Amazon, and is it compatible with my projector?
How big is the cheapest collapsible bike, and will it fit in my car boot? Find me the best coffee... and tell me more about the flavour."
The Product Advisor Question Answering team‘s role is to empower Alexa to give the best possible answers to our customers' product questions. Where we have multiple answers, we need to differentiate and rank between them, based on nuanced criteria.
We then need to present a clear and specific answer to the customer’s question, in natural language, providing a true shopping assistant experience - with augmented visuals and media for our GUI Alexa devices. And we need to constantly measure our output to customers, and quantify our continuous improvement, for each and every customer question.
To meet this challenge, we need to work on ambiguous problems cutting across many technical and business domains, and leverage a wide variety of technology, and sources of product data, from Amazon and third parties.
Natural language understanding, speech to text conversion, Web APIs and services, data stores, knowledge graphs, ML solutions, search indexes, the web – no stone can be left unturned.
You will be joining an experienced, agile team, with high standards of quality and operational excellence, working on a modern codebase.
We have a continuous deployment pipeline, monitoring for peer review, static analysis, automated testing and coverage at various levels (unit, mutation, integration, end to end, load, performance), automated rollback, phased deployments, A/B testing and testing in production.
Join us, and help build the world’s best personal shopping assistant.