We are looking for a Data Scientist to join the Devices Demand Planning team that manages forecasting and optimization for the entire Amazon device family of products and accessories. Predicting the sales volume for every Amazon device and accessory for every day of the year 6 to 12 months in advance requires the ability to develop and use scalable and robust state-of-the-art algorithms that involve learning from large amounts of data, such as customer engagement data (impression, clicks, transactions, etc.), product features (product attributes, price, promotion, etc.), merchandising activities, relevant products and users, in order to drive more efficient customer engagement and business values.
As our organization continues to grow, our new ML platforms are being put into place to make discovery, prototyping and delivery of new algorithms and features faster in an environment that requires rapid innovation and expertise.
This role is central to the continued growth of the Amazon Device division as we continue to drive sales, reduce costs and become more central as Amazon continues to find ways to increase profits, reduce waste, and maintain our position as on the leading edge of technology. We have grown from only supporting the first Kindle e-reader to a vast portfolio of Fire tablets, Fire TVs, Echo, and Dash buttons. You will have an opportunity to both develop advanced scientific solutions, drive critical customer and business impacts, and work closely with the development team to pave the way for our team, and all of Amazon. You will play a key role to drive end-to-end solutions from understanding our business and business requirements, identifying opportunities from a large amount of historical data, building prototypes and exploring conceptually new solutions, running online experiments, to working with partner teams for prod deployment. You will collaborate closely with engineering peers as well as business stakeholders. You will be at the heart of a growing and exciting focus area for Amazon Devices.
In a typical day, you will work closely with talented machine learning scientists, statisticians, software engineers, and business groups. Your work will include cutting edge technologies that enable implementation of sophisticated models on big data. As a successful data scientist in our Demand Forecasting team, you are an analytical problem solver who enjoys diving into data, is excited about investigations and algorithms, can multi-task, and can credibly interface between technical teams and business stakeholders. Your analytical abilities, business understanding, and technical savvy will be used to identify specific and actionable opportunities to solve existing business problems in Demand Forecasting, through collaboration with engineering, research, and business teams. Your expertise in synthesizing and communicating insights and recommendations to audiences of varying levels of technical sophistication will enable you to answer specific business questions and innovate for the future.
Major responsibilities include:
· Translating Demand Forecasting business questions and concerns into specific analytical questions that can be answered with available data using statistical and machine learning methods; working with engineers to produce the required data when it is not available
· Providing feedback to our science and engineering teams on the applicability of technical solutions from the business perspective
· Presenting critical data in a format that is immediately useful to answer questions about the inputs and outputs of Demand Forecasting systems and improving their performance
· Communicating verbally and in writing to business customers with various levels of technical knowledge, educating them about our systems, as well as sharing insights and recommendations
· Improving upon existing Demand Forecasting statistical or machine learning methodologies by developing new data sources, testing model enhancements, running computational experiments, and fine-tuning model parameters for new forecasting models
· Supporting decision making by providing requirements to develop analytic capabilities, platforms, pipelines and metrics then using them to analyze trends and find root causes of forecast inaccuracy
· Formalizing assumptions about how demand forecasts are expected to behave, creating definitions of outliers, developing methods to systematically identify these outliers, and explaining why they are reasonable or identifying fixes for them
· Utilizing code (Python, R, Scala, etc.) for analyzing data and building statistical and machine learning models and algorithms