Data Scientist Amazon
Newark, NJ

Job Description

Audible, an Amazon company, is the leading provider of premium digital spoken audio information and entertainment on the internet, offering customers a new way to enhance and enrich their lives every day.

Audible Data Scientists are members of an interdisciplinary research team with an integral role in the design and integration of models to automate decision making throughout the business in every country. We empower the cutting-edge machine learning and deep learning techniques in the many areas of the business, including but not limited to Customer: segmentation, acquisition, retention, engagement; Product: customer experience optimization, simulation, testing and evaluation, Content: recommender system, natural language processing in text and voice, and International business etc. We translate business goals into agile, insightful analytics and seek to create value for both stakeholders and customers and deliver findings in a clear and actionable way.

We are currently looking for a Data Scientist to support the ongoing development of Product vertical. The candidate will be expected to work closely with our team members on the design, development, deployment of ML/NLP/DS models which built upon the cutting-edge technologies. Additionally, we are seeking candidates with strong rigor in data sciences, engineering, creativity, curiosity, and great judgment.

KEY RESPONSIBILITIES

* Identifying necessary, relevant, and novel data sources and acquiring data, which often means building the necessary SQL / ETL queries, import processes through various company specific interfaces for accessing Red Shift storage systems, but also by building relationships with stakeholders and counterparts all over the world in order to form trusting, functional relationships that provide for a sustainable flow and sharing of information.
* Exploring data will occupy the largest portion of attention, and should be second nature in order to deeply understand the phenomenon being modeled, and the validity and reliability of the inputs, including but not limited to inspecting univariate distributions, exploring bivariate relationships, constructing appropriate transformations, and tracking down the source and meaning of anomalies when and where they arise.
* Model building should draw from any approach that enhances accuracy and understanding including statistical modeling, mathematical modeling, network modeling, social network modeling, natural language processing, machine learning, algorithms, genetic algorithms, and neural networks.
* Validating models against alternative approaches, expected and observed outcome, and numerous directly and indirectly relevant business defined key performance indicators.
* Reviewing models of peers for the purpose of reducing and managing risk to the business, and maximizing improvement of business practice and customer experience.
* Implementing models from the initial evaluation of the computational demands, accuracy, and reliability of the relevant ETL processes, and the integrity of the data sources in production, to the computational demands, accuracy, and reliability of the simpler scoring processes, to the computational demands, accuracy, and reliability of model training in higher and higher frequencies in the production environment.
* Operationalization will include identifying the stakeholders in the life of a model score throughout the company and around the world, forming trusting bonds, understanding the phenomenon, optimizing the model output for integration to the practice, assuring "buy in" from practitioners as well as leaders, training for proper utilization and communication, and assessing proper utilization.
* Model management will include developing sustainable, consumable, accurate, and impactful reporting on model inputs, model outputs, observed outputs, business impact, and key performance indicators.
* Data scientists must be able to discuss their research in any level of detail with their peers, and with appropriate calibration to stakeholders in small and large group settings. They are expected to acquire support and partnership from personalities through the company, including engineering and research teams at Amazon globally. Successful data scientists are expected to influence and mentor each other, influence and mentor their peers globally, and influence and mentor their partners in the business in every country.