<strong>Description<br><br></strong>Orchestrating the selection of one out of tens of millions of ads, honoring advertiser targeting intent for hundreds of thousands of advertisers while ensuring great shopper experience for billions of shoppers millions of times per second on a latency of tens of milliseconds is not a trivial task. The demand retrieval team within the Amazon DSP organisation deals with this challenge, developing and operating machine learning models that match ads opportunities with the most relevant ads to deliver the right messages to the right customers at the right time.<br><br>We are looking for an Applied Scientist to optimize ad matching for Amazon’s programmatic advertisement products. In this role you will lead the design and implementation of solutions for performance sourcing, using behavioural information on customers’ interactions with Amazon and other owned and operated businesses as well as contextual information about the bid request to predict their propensity to convert, in turn driving better advertising campaign outcomes. Your work will affect multi-billion dollar businesses, and you will be responsible for designing, testing and delivering significant breakthrough's for Amazon's business.<br><br>Successful candidates will have strong technical ability, excellent teamwork, communication skills, and a motivation to achieve business results in a fast-paced environment.<br><br>Key job responsibilities<br><br><ul><li> Design and implement deep learning models to match the right customers with the right ads across different verticals, geographies, and ads formats.</li><li> Investigate new ML techniques such as multi-task learning to ensure that models can operate for a variety of advertisers in multiple industries and with different volumes of conversion events.</li><li> Improve the performance, generalisation and scalability of models by introducing new features and enhancing models’ architecture.</li><li> Work side by side with our engineers to deliver code changes impacting our ads stack, working with very large datasets and high throughput production systems.</li><li> Rapidly prototype and test many possible hypotheses/implementation alternatives in a high-ambiguity environment, making use of both quantitative analysis and business judgement.</li><li> Be immersed in Amazon's advertisers and their objectives, and think long-term about how to turn those objectives into products and technical capabilities.</li><li> Understand the latest literature on machine learning for recommender and advertising systems, contributing to guiding strategic investment for the organization.<br><br></li></ul>A day in the life<br><br>You will partner with our product and engineering teams, bringing your own ideas to the conversation and aligning on work, adjusting priorities based on business requirements and fast iteration on experiments. You will have a strong theoretical understanding of modern ML techniques and methodologies, and the software engineering and data processing skills to deploy these using the large-scale datasets we deal with in advertising.<br><br><strong>About The Team<br><br></strong>The Demand Retrieval team is responsible for designing, implementing, deploying and operating machine learning models that match bid opportunities to ads demand based on performance, campaign delivery, and targeting objectives specified by advertisers. We measure the success of our approaches based on offline experimentation and and online metrics that measure the impact of our matching models on campaign KPIs (e.g.: cost per action, return on ads investment, budgets delivered, and targeting precision).<br><br><strong>Basic Qualifications<br><br></strong><ul><li> 3+ years of building machine learning models for business application experience</li><li> PhD, or Master's degree and 6+ years of applied research experience</li><li> Experience programming in Java, C++, Python or related language</li><li> Experience with neural deep learning methods and machine learning<br><br></li></ul><strong>Preferred Qualifications<br><br></strong><ul><li> Experience with modeling tools such as R, scikit-learn, Spark MLLib, MxNet, Tensorflow, numpy, scipy etc.</li><li> Experience with large scale distributed systems such as Hadoop, Spark etc.<br><br></li></ul>Amazon is an equal opportunity employer and does not discriminate on the basis of protected veteran status, disability, or other legally protected status.<br><br>Our inclusive culture empowers Amazonians to deliver the best results for our customers. If you have a disability and need a workplace accommodation or adjustment during the application and hiring process, including support for the interview or onboarding process, please visit for more information. If the country/region you’re applying in isn’t listed, please contact your Recruiting Partner.<br><br>The base salary range for this position is listed below. Your Amazon package will include sign-on payments and restricted stock units (RSUs). Final compensation will be determined based on factors including experience, qualifications, and location. Amazon also offers comprehensive benefits including health insurance (medical, dental, vision, prescription, Basic Life & AD&D insurance and option for Supplemental life plans, EAP, Mental Health Support, Medical Advice Line, Flexible Spending Accounts, Adoption and Surrogacy Reimbursement coverage), 401(k) matching, paid time off, and parental leave. Learn more about our benefits at NY, New York - 183,800.00 - 248,700.00 USD annually<br><br><br><strong>Company</strong> - Amazon Development Center U.S., Inc.<br><br>Job ID: A10540955