We are looking for an inquisitive, life long learner with a keen interest on evidence driven decision making, e-commerce and digital businesses and understanding of customer behavior. We are seeking a researcher with interest in e-commerce industry and relevant experience to leverage and help evolve our marketing experimentation frameworks. Our candidate will have strong causal inference, experimental design, decision theory and bayesian statistics background. This is a role to flesh out a nascent new program that aspires to be the beacon for the industry. So this is genuinely a unique opportunity for candidates to emerge as world class experts in a fast evolving, ever green domain. Our mission is to engage customers with the right products and services to enable a great shopping experience. You will go home and show your family and friends why they receive this ad on search or social channels or that email from Amazon. You will make a difference by improving the relevancy for customers and optimizing the investment level for Amazon. Cutting edge technology and algorithms including econometric methods, statistical modeling, machine learning, and data mining are the core of our business. Marketing drives a large portion of Amazon’s traffic and business, and represents a unique opportunity to drive impact on the company’s bottom line. We also focus on developing novel A/B experimentation mechanisms to measure efficacy of our ML solutions. With essentially full ownership of our own product roadmap, there is a large R&D component to our work with together with sound business understanding and an appetite for innovation are highly valued. Key job responsibilities - Design and build scalable analytic solutions using statistical models to measure the business impact of cross channel marketing treatments. - Leverage causal inference methods to estimate the impact of large-scale investments. - Work closely with both business units and engineering teams to formulate business problems in experimental designs and associated technical solution strategies. - Develop a library of measurement solutions to enable data-driven decision making. - Provide inputs to the product roadmap, emphasizing research and development (R&D) to continuously enhance bidding capabilities. - Superior verbal and written communication and presentation skills, ability to convey rigorous mathematical concepts and considerations to non-experts. A day in the life As a Economist on our team, you will leverage you strong background in statistics and causal inference to help build the next generation of marketing experimentation frameworks. This role requires a pragmatic technical leader comfortable with ambiguity, capable of summarizing complex data and models through clear visual and written explanations. The ideal candidate will have experience with causal inference, experimental design, decision theory and bayesian statistics. We are particularly interested in experience in building large scale marketing experimentation frameworks. About the team We are a team of scientists who are set to build solutions in production. We work on prediction, optimization, and experimentation problems to provide data-driven inputs to marketing decisions and build highly scalable statistical frameworks and machine learning models across Automated Marketing and Events (AME) org to drive long-term profitability. Specifically, the team focuses on building re-usable science solutions to address three focal areas: (i) Content selection, creation and moderation, (ii) Bidding which involves valuation, efficiency management and net-profit maximization via elasticity measurement, and (iii) Scalable Experimentation frameworks and statistical techniques for designing and performing causal analysis. efficiency management and net-profit maximization via elasticity measurement, and (iii) Scalable Experimentation frameworks and statistical techniques for designing and performing causal analysis.
April 30, 2024Using causal random forests and Bayesian structural time series to extrapolate from sparse data ensures that customers get the most useful information as soon as possible.
