Applied Scientist II - Community Shopping (NLP) Amazon
Seattle, WA

Job Description

Our team is mining for interesting topics and the authoritative relationships between them. We blend of a variety of disciplines (such as NLP, data mining, machine learning, image classification, big data, semantic web, graph stores, cloud computing) in an effort to understand our customers and the things they're excited bout. For example: "Harry Potter" is a fictional universe based around a series of books, and "Harry Potter Wands" are essential costume components. While there are knowledge and entity graphs out there, none capture the rich lifestyles and topics our customers are interested in.

We are a new team dedicated to understanding What Makes Stuff (And People) Awesome. Amazon customers are people with hobbies, passions, and quirky interests. They run barefoot marathons, brew beer, cry during Disney movies, and cosplay Sith characters at Comic-Con. We're looking for inquisitive developers who want to help us understand the world and the people in it.

How do you build software that understands things? How do you know who loves what? Did Joe Customer buy a blender because a) he loves whole-food smoothies b) he's equipping his college apartment or c) his cheap college blender burst into flames and nearly burned down the dorm? Answering those questions is exciting - especially the pyrotechnic blender part - and that's why we think this is one of the most interesting teams at Amazon.

If you'd want to make things about 20% cooler, think bowties are cool, would just as soon kiss a Wookie as work somewhere boring (or if you know which fictional universes these quotes are from), give the team a shout. We know submitting online applications can be impersonal. So feel free to contact us directly to introduce yourself and learn more: hobbyt-jobs (@) amazon.com.

About you:

You're a seasoned scientist looking for a career where you'll be able to innovate, to deliver, and to influence. You are passionate about start-up opportunities, where you succeed by fearlessly testing new ideas and rapidly iterating on technical solutions. You have a demonstrable track record of success delivering high quality software products and services at scale. You are a coding ninja with strong OO design skills, knowledge of Machine Learning and distributed system architectures based on SOA. You are a team guru who leads and mentors junior scientists and consults product owners.

About us together:

We're going to change the way that shoppers discover and evaluate products online. We will work together on defining features and our overall product roadmap. You will be a technical leader who owns core services and user applications. You will drive system architecture and establish sound software development practices. You will be writing code, participating in code reviews and creating documentation. We will dive deep into customer contributed data and system operational metrics identifying areas of improvement. We'll make tough decisions together, but we'll all understand why. We'll be the dream team.

Some technologies and problem spaces we'll be working on:

HIGHLY SCALABLE DISTRIBUTED SYSTEMS - Amazon's data has billions of records for a mind-boggling variety of products and millions of customers in multiple countries. We will leverage Amazon's cloud infrastructure to build systems and design algorithms that are able to handle these large amounts of data. Where traditional solutions fail we develop approximate, distributed, and streaming algorithms.

DATA MINING AND NATURAL LANGUAGE PROCESSING - We will use the latest machine learning and predictive modeling technologies to extract information valuable to shoppers. We will look for patterns, train thousands of models, and summarize data to gain insights about customer product needs and experiences. We continually evaluate the effectiveness of our solutions and ways to improve them.

If you want to design, develop and implement systems at scale now and are looking for a challenge, now is a great time to talk to the Hobbyt team!