Classical Machine Learning Amazon.png

The history of Amazon's recommendation algorithm

Collaborative filtering and beyond.

In 2017, when the journal IEEE Internet Computing was celebrating its 20th anniversary, its editorial board decided to identify the single paper from its publication history that had best withstood the “test of time”. The honor went to a 2003 paper called “ Recommendations: Item-to-Item Collaborative Filtering”, by then Amazon researchers Greg Linden, Brent Smith, and Jeremy York.

Collaborative filtering is the most common way to do product recommendation online. It’s “collaborative” because it predicts a given customer’s preferences on the basis of other customers’.

“There was already a lot of interest and work in it,” says Smith, now the leader of Amazon’s Weblab, which does A/B testing (structured testing of variant offerings) at scale to enable data-driven business decisions. “The world was focused on user-based collaborative filtering. A user comes to the website: What other users are like them? We sort of turned it on its head and found a different way of doing it that had a lot better scaling and quality characteristics for online recommendations.”

The better way was to base product recommendations not on similarities between customers but on correlations between products. With user-based collaborative filtering, a visitor to would be matched with other customers who had similar purchase histories, and those purchase histories would suggest recommendations for the visitor.

With item-to-item collaborative filtering, on the other hand, the recommendation algorithm would review the visitor’s recent purchase history and for each purchase, pull up a list of related items. Items that showed up repeatedly across all the lists were candidates for recommendation to the visitor. But those candidates were given greater or lesser weight depending on how related they were to the visitors’ prior purchases.

That notion of relatedness is still derived from customers’ purchase histories: item B is related to item A if customers who buy A are unusually likely to buy B as well. But Amazon’s Personalization team found, empirically, that analyzing purchase histories at the item level yielded better recommendations than analyzing them at the customer level.

Family ties

Beyond improving recommendations, item-to-item collaborative filtering also offered significant computational advantages. Finding the group of customers whose purchase histories most closely resemble a given visitor’s would require comparing purchase histories across Amazon’s entire customer database. That would be prohibitively time consuming during a single site visit.

The alternatives are either to randomly sample other customers in real time and settle for the best matches found or to build a huge offline similarity index by comparing every customer to every other. Because Amazon customers’ purchase histories can change dramatically in the course of a single day, that index would have to be updated regularly. Even offline indexing presents a huge computational burden.

Consumer division CEO Jeff Wilke discusses the history of Amazon's recommendation algorithm
Excerpt from keynote at re:MARS 2019

On average, however, a given product sold on is purchased by only a tiny subset of the site’s customers. That means that inspecting the recent-purchase histories of everyone who bought a given item requires far fewer lookups than identifying the customers who most resemble a given site visitor. Smith and his colleagues found that even with early-2000s technology, it was computationally feasible to produce an updated list of related items for every product on the Amazon site on a daily basis.

The crucial question: how to measure relatedness. Simply counting how often purchasers of item A also bought item B wouldn’t do; that would make a few bestsellers like Harry Potter books and trash bags the top recommendations for every customer on every purchase.

Instead, the Amazon researchers used a relatedness metric based on differential probabilities: item B is related to item A if purchasers of A are more likely to buy B than the average Amazon customer is. The greater the difference in probability, the greater the items’ relatedness.

When Linden, Smith, and York published their paper in IEEE Internet Computing, their item-based recommendation algorithm had already been in use for six years. But it took several more years to identify and correct a fundamental flaw in the relatedness measure.

Getting the math right

The problem: the algorithm was systematically underestimating the baseline likelihood that someone who bought A would also buy B. Since a customer who buys a lot of products is more likely to buy A than a customer who buys few products, A buyers are, on average, heavier buyers than the typical Amazon customer. But because they’re heavy buyers, they’re also unusually likely to buy B.

Smith and his colleagues realized that it wasn’t enough to assess the increased likelihood of buying product B given the purchase of product A; they had to assess the increased likelihood of buying product B with any given purchase. That is, they discounted heavy buyers’ increased likelihood of buying B according to the heaviness of their buying.

“That was a large improvement to recommendations quality, when we got the math right,” Smith says.

That was more than a decade ago. Since then, Amazon researchers have been investigating a wide variety of ways to make customer recommendations more useful: moving beyond collaborative filtering to factor in personal preferences such as brands or fashion styles; learning to time recommendations (you may want to order more diapers!); and learning to target recommendations to different users of the same account, among many other things.

In June 2019, during a keynote address at Amazon’s first re:MARS conference, Jeff Wilke, the CEO of Amazon’s consumer division, highlighted one particular advance, in the algorithm for recommending movies to Amazon’s Prime Video customers. Amazon researchers’ innovations led to a twofold improvement in that algorithm’s performance, which Wilke described as a “once-in-a-decade leap”.

Entering the matrix

Recommendation is often modeled as a matrix completion problem. Imagine a huge grid, whose rows represent Prime Video customers and whose columns represent the movies in the Prime Video catalogue. If a customer has seen a particular movie, the corresponding cell in the grid contains a one; if not, it’s blank. The goal of matrix completion is to fill in the grid with the probabilities that any given customer will watch any given movie.

In 2014, Vijai Mohan’s team in the Personalization group — Avishkar Misra, Jane You, Rejith Joseph, Scott Le Grand, and Eric Nalisnick — was asked to design a new recommendation algorithm for Prime Video. At the time, the standard technique for generating personalized recommendations was matrix factorization, which identifies relatively small matrices that, multiplied together, will approximate a much larger matrix.

Inspired by work done by Ruslan Salakhutdinov — then an assistant professor of computer science at the University of Toronto — Mohan’s team instead decided to apply deep neural networks to the problem of matrix completion.

The typical deep neural network contains thousands or even millions of simple processing nodes, arranged into layers. Data is fed into the nodes of the bottom layer, which process it and pass their results to the next layer, and so on; the output of the top layer represents the result of some computation.

Training the network consists of feeding it lots of sample inputs and outputs. During training, the network’s settings are constantly adjusted, until they minimize the average discrepancy between the top layer’s output and the target outputs in the training examples.


Matrix completion methods commonly use a type of neural network called an autoencoder. The autoencoder is trained simply to output the same data it takes as input. But in-between the input and output layers is a bottleneck, a layer with relatively few nodes — in this case, only 100, versus tens of thousands of input and output nodes.

As a consequence, the network can’t just copy inputs directly to outputs; it must learn a general procedure for compressing and then re-expanding every example in the training set. The re-expansion will be imperfect: in the movie recommendation setting, the network will guess that customers have seen movies they haven’t. But when, for a given customer-movie pair, it guesses wrong with high confidence, that’s a good sign that the customer would be interested in that movie.

To benchmark the autoencoder’s performance, the researchers compared it to two baseline systems. One was the latest version of Smith and his colleagues’ collaborative-filtering algorithm. The other was a simple listing of the most popular movie rentals of the previous two weeks. “In the recommendations world, there’s a cardinal rule,” Mohan says. “If I know nothing about you, then the best things to recommend to you are the most popular things in the world.”

To their mild surprise, the item-to-item collaborative-filtering algorithm outperformed the autoencoder. But to their much greater surprise, so did the simple bestseller list. The autoencoder’s performance was “so bad that we had to go and doublecheck and re-run the experiments multiple times,” Mohan says. “I was giving a hard time to the scientists. I was saying, ‘You probably made a mistake.’”

Once they were sure the results were valid, however, they were quick to see why. In a vacuum, matrix completion may give the best overview of a particular customer’s tastes. But at any given time, most movie watchers will probably opt for recent releases over neglected classics in their preferred genres.

Neural network classifiers with time considerations
Amazon researchers found that using neural networks to generate movie recommendations worked much better when they sorted the input data chronologically and used it to predict future movie preferences over a short (one- to two-week) period.

So Mohan’s team re-framed the problem. They still used an autoencoder, but they trained it on movie-viewing data that had been sorted chronologically. During training, the autoencoder saw data on movies that customers had watched before some cutoff time. But it was evaluated on how well it predicted the movies they had watched in the two-week period after the cutoff time.

Because Prime Video’s Web interface displays six movie recommendations on the page associated with each title in its catalogue, the researchers evaluated their system on whether at least one of its top six recommendations for a given customer was in fact a movie that that customer watched in the two-week period after the cutoff date. By that measure, not only did the autoencoder outperform the bestseller list, but it also outperformed item-to-item collaborative filtering, two to one. As Wilke put it at re:MARS, “We had a winner.”

Whether any of the work that Amazon researchers are doing now will win test-of-time awards two decades hence remains to be seen. But Smith, Mohan, and their colleagues will continue to pursue new approaches to designing recommendation algorithms, in the hope of making that much more useful for customers.

Work with us

See More Jobs
US, WA, Seattle
MULTIPLE POSITIONS AVAILABLEEntity: Services LLCTitle: Applied Scientist IIWorksite: Seattle, WAPosition Responsibilities:Participate in the design, development, evaluation, deployment and updating of data-driven models and analytical solutions for machine learning (ML) and/or natural language (NL) applications. Develop and/or apply statistical modeling techniques (e.g. Bayesian models and deep neural networks), optimization methods, and other ML techniques to different applications in business and engineering. Routinely build and deploy ML models on available data. Research and implement novel ML and statistical approaches to add value to the business. Mentor junior engineers and is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation #0000
US, WA, Seattle
Do you want to join Alexa AI -- the science team behind Amazon’s intelligence voice assistance system? Do you want to utilize cutting-edge deep-learning and machine learning algorithms to delight millions of Alexa users around the world?If your answers to these questions are “yes”, then come join us at the Alexa Artificial Intelligence team, which is in charge of improving Alexa user satisfaction through real-time metrics monitoring and continuous closed-loop learning. The team owns the modules that reduce user perceived defects and frictions through utterance reformulation, contextual and personalized hypothesis ranking.With the Alexa Artificial Intelligence team, you will be working alongside a team of experienced machine/deep learning scientists and engineers to create data driven machine learning models and solutions on tasks such as sequence-to-sequence query reformulation, graph feature embedding, personalized ranking, etc..You will be expected to:· Analyze, understand, and model user-behavior and the user-experience based on large scale data, to detect key factors causing satisfaction and dissatisfaction (SAT/DSAT).· Build and measure novel online & offline metrics for personal digital assistants and user scenarios, on diverse devices and endpoints· Create and innovate deep learning and/or machine learning based algorithms for utterance reformulation and contextual hypothesis ranking to reduce user dissatisfaction in various scenarios;· Perform model/data analysis and monitor user-experienced based metrics through online A/B testing;· Research and implement novel machine learning and deep learning algorithms and models.
US, WA, Seattle
MULTIPLE POSITIONS AVAILABLEEntity: Amazon Web Services, Inc.Job Title: Data Scientist IILocation: Seattle, WAPosition Responsibilities:Manage metrics reporting, perform data mining and big data analysis, to provide strategic advice on business and improve processes and systems. Collect business use cases, research and evaluate opportunities, and leverage data to support business functions through advanced mathematical modeling. Independently own and solve business problems. Transform business problems into mathematical models, provide data-driven solutions, and write clear and factually correct documents explaining technical concept to non-technical audience. Gather and use complex data sets across domains and proactively gather data when not readily available. Use specialized modelling software and scripting languages to scale methods. Partner with business, science, and engineering to establish scalable, efficient, automated processes for large scale data analyses, model development, model validation, and model is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation #0000
US, WA, Seattle
Workforce Staffing (WFS) is responsible for filling all of our First Mile, Middle Mile, and Last Mile Operations with hourly labor. In 2019, WFS hired hundred of thousands of hourly associates across NA and EU and will receive over millions of job applications for employment. This role is part of the Workforce Intelligence Team, tasked acquiring, modeling and visualizing all the data required to report out on performance metrics such as fill rates and funnel statistics, and forecasting hiring volumes to predict hiring risks and to support internal capacity planning.As part of the Workforce Intelligence team, the Ops Research Scientist will work on forecasting and optimization projects to improve funnel efficiency. The Ops Research Scientist will partner with capacity planning leaders, product/program managers, data engineers and other research scientists to build tools with clear business impact in an exciting and fast-paced start-up environment. The OR Scientist will be responsible for developing new predictive and optimization models as well as algorithms for key applications. The OR Scientist will be expected to be a thought leader as we chart new courses with our labor planning models. Successful candidates will have a deep knowledge of computational optimization methods and mathematical modeling, background in statistical and machine learning methods, the ability to map models into production-worthy code, the communication skills necessary to explain complex technical approaches to a variety of stakeholders and customers, and the excitement to take iterative approaches to tackle big, long term problems.Key Responsibilities:· Guide the technical approach for the design and implementation of successful models and algorithms in support of teams across Amazon.· Work in expert cross-functional teams delivering on demanding projects.· Functionally decompose complex problems into simple, straight-forward solutions.· Share knowledge in state-of-the-art statistics and machine learning research or frontier applied mathematical modeling and computation applicable to our problem space.
US, WA, Seattle
Within Amazon’s Corporate Financial Planning & Analysis team (FP&A), we enjoy a unique vantage point into everything happening within Amazon. As part of that, this role would be part of a team that is responsible for Company’s enterprise-wide financial planning & analytics environment.- Are you excited about working directly to empower users?- Love to get your hands dirty and solve challenging technical issues?We are looking for a customer obsessed Data Scientist who can apply the latest research, state of the art algorithms and machine learning to build highly scalable systems in the financial planning and analytics domain.The successful candidate will have strong data mining and modeling skills and is comfortable facilitating and working from concept through to execution. This role will also build tools and support structures needed to analyze data and present findings to business partners to drive improvements.The data flowing through our platform directly contributes to decision-making by our CFO and all levels of finance leadership. If you’re passionate about building tools that enhance productivity, improve financial accuracy, reduce waste, and improve work-life harmony for a large and rapidly growing finance user base, come join us! If you are passionate about solving complex problems, in a challenging environment, we would love to talk with you.Responsibilities of this position include:A qualified candidate must have demonstrated ability to manage modeling projects, identify requirements and build methodology and tools that are statistically grounded but also explainable operationally, apply technical skills allowing the models to adapt to changing attributes. In addition to the modeling and technical skills, possess strong written and verbal communication skills, strong focus on internal customers, and high intellectual curiosity with ability to learn new concepts/frameworks, algorithms and technology rapidly as changes arise.This is a tremendous opportunity to develop creative, new and innovative ways of interacting with and interpreting business key performance indicators.We're looking for innovators who want to create reliable and scalable solutions to help our leaders create the appropriate strategies.Additional responsibilities may include:· Research machine learning algorithms and implement by tailoring to particular business needs and tested on large datasets.· Predict future customer behavior and business conditions (Machine Learning, Predictive Modeling) and manipulating/mining data from database tables (Redshift, Oracle, Data Warehouse)· Create automated metrics· Providing analytical network support to improve quality and standard work results· Root cause research to identify process breakdowns within departments and providing data through use of various skill sets to find solutions to breakdown· Foster culture of continuous improvement for Customer Experience through all aspects of Data;(BI, Data Analytics, and Data Science)· Participate in the full development life cycle, end-to-end with cross functional teams, from design, implementation and testing, to documentation, delivery, support, and maintenance.As a member of our team you will be responsible for modelling complex problems, discovering insights and identifying opportunities through the use of statistical, machine learning, algorithmic, data mining and visualization techniques. You will need to collaborate effectively with internal stakeholders and cross-functional teams to solve problems, and create operational efficiencies. You should be able to apply a breadth of tools, data sources and analytical techniques to answer a wide range of high-impact business questions and present the insights in concise and effective manner. Additionally, you will need to be entrepreneurial, able to deal with high ambiguity and should be an effective communicator capable of independently driving issues to resolution and communicating insights to technical and non-technical audiences.
US, WA, Seattle
Voice-driven AI experiences are finally becoming a reality and Amazon’s Alexa voice cloud service and Echo devices are at the forefront of this latest technology wave. We deliver world-class products on aggressive schedules that are used every day, by people you know, in and about their homes. At the same time, we obsess about customer trust and ensure that we build products in a manner that maintains our high bar for customer privacy. We are looking for a passionate and talented data science leader to develop the strategies for understanding where accuracy gaps exist today, drive new scientific techniques to address them short- and long-term plan and establish the mechanisms for a successful execution.This is a unique opportunity to play a key role in an exciting, fast growing business. You come with a start-up mentality and skills that span research science, software engineering, product and strategic vision. You will work closely with talented engineers and lead ML scientists to put the algorithms and models into practice. Your work will directly impact the trust customers place in Alexa, globally.You are the ideal candidate if you are clearly passionate about delivering experiences that delight customers and creating solutions that are robust. You should thrive in ambiguous environments that require to find solutions to problems that have not been solved before. You enjoy and succeed in fast paced environments where learning new concepts quickly is a must. You leverage your exceptional technical expertise, a sound understanding of the fundamentals of Computer Science, and practical experience building large-scale distributed systems to creating reliable, scalable, and high performance products. You provide technical and scientific guidance to your team members. Your strong communication skills enable you to work effectively with both business and technical partnersYou will be joining a select group of people making history producing one of the most highly rated products in Amazon's history.
US, WA, Seattle
Work at the intersection of data science and economics.The DAC AdsEcon Team is looking for a Data Scientist II to help and be part of a team to put cutting edge economic and data science advertising research into production. We are looking for a unique individual who is interested in bigger picture strategic thinking but with the passion for big data.Advertising is used daily to surface new selection and provide customers a wider set of product choices along their shopping journeys. The business is focused on generating value for shoppers as well as advertisers. Our team uses econometrics, machine learning, and data science to help advertisers choose the right advertising product to meet their marketing goals. We also generate insights to guide Amazon Advertising strategy, providing direct support to the high level leaders.If you have a background in economics, computer science, statistics, or mathematics and have a passion for solving large, and impactful problems, this is the job for you. Key responsibilities of Data Scientist include the following:· Partnering with economists and senior team members to drive science improvements and implement technical solutions at the cutting edge of machine learning and econometrics· Helping build data systems that leverage diverse data sources to understand how different advertiser’s decisions impact their performance across multiple advertising products.· Build interpretable statistical models and analyze experiment results to answer questions that will drive high impact decisions across Amazon.About Amazon's Advertising business:Amazon is investing heavily in building a world class advertising business and we are responsible for defining and delivering a collection of self-service performance advertising products that drive discovery and sales. Our products are strategically important to our Retail and Marketplace businesses driving long term growth. We deliver billions of ad impressions and millions of clicks daily and are breaking fresh ground to create world-class products. We are highly motivated, collaborative and fun-loving with an entrepreneurial spirit and bias for action. With a broad mandate to experiment and innovate, we are growing at an unprecedented rate with a seemingly endless range of new opportunities.
US, WA, Seattle
About Prime Video: Prime Video is changing the way people watch movies and TV shows, offering more than 150,000 new release movies, next-day television shows and classic favorites available to rent or purchase on-demand, and more than 38,000 titles available to customers with an Amazon Prime membership. We believe so deeply in the mission of our video offering that we've launched our own Amazon Studios to create Original and Exclusive content. With an Amazon Prime membership, customers can have unlimited access to thousands of titles for no additional charge, including popular and award-winning Prime Originals like Jack Ryan, Fleabag and The Marvelous Mrs. Maisel.About the team: The vision for the engagement automation team is to inspire our customers to engage with all that Prime Video brand has to offer. To achieve our vision, we create product and technology solutions that drive incremental activation and engagement of PV customers worldwide. We obsess over finding effective ways to reach active and inactive customers with relevant and timely content that drives traffic to the PV experience. Using smart rules and machine learning we generate relevant, timely, and personalized engagement opportunities via a broad portfolio of both in-app and out-of-app experiences, on a fully automated basis.About the role: We seek an experienced Applied Scientist to be join Engagement Automation. Join us in defining and designing a fully automated E2E engagement system powered by science to increase customer engagement, activation and global adaption of prime video. You will have the opportunity to apply latest neural network, deep learning, and transfer learning models to define the target audience for Prime Video Originals and shows. You will also have the ability to apply causal modeling to identify customers with the most incremental impact from marketing activity as well as the defining characteristics of the most effective marketing campaigns. Additionally there is opportunity to apply Reinforcement Learning techniques to define the optimal marketing strategy (frequency, recency, touch point, channel, creative, copy) for each customer.You should expect to exercise both your coding skills and creative research thinking as you map real world processes to ML enabled systems. You will spend your time as a hands-on machine learning practitioner and a research leader. You will play a key role on the team, building and guiding machine learning models from the ground up. If you’re looking for an opportunity to make a big impact in a global business with a startup culture, we’re looking for you.
US, VA, Arlington
Amazon’s Talent Assessment team designs and implements hiring systems for one of the world’s fastest growing companies. We work in a data-focused, global environment solving complex problems with deep thought, large-sample research, and advanced quantitative methods to deliver practical solutions that make hiring more fair, accurate, and efficient.We're looking for an experienced assessment and personnel selection scientist who is equal parts researcher, consultant, and thought leader, with strong expertise in psychometrics, research methodology, and data analysis. In this role, you will collaborate with cross-functional teams of psychologists, ML scientists, UX researchers, engineers, and product managers, to direct the research, development, and implementation of new assessment methods to measure exactly what it requires to be an engaged and successful employee at Amazon.What you’ll do:· Lead the development and research of new content and approaches to assessment (e.g., high fidelity simulation, non-cognitive computer adaptive testing, interactive item types)· Design and execute large-scale, highly-visible global assessment validation and optimization projects· Develop assessment content, including personality, cognitive ability, and simulations· Perform complex statistical/quantitative analyses with large datasets· Apply the scientific method to answer novel research questions· Influence executive project sponsors and stakeholders across the company· Drive effective teamwork, communication, collaboration and commitment across cross-functional groups with competing priorities
US, WA, Seattle
The Economic Technology team (ET) is looking for a Senior Applied Scientist to join our team in building Reinforcement Learning solutions at scale. The ET applies Machine Learning, Reinforcement Learning, Causal Inference, and Econometrics/Economics to derive actionable insights about the complex economy of Amazon’s retail business. We also develop Statistical Models and Algorithms to drive strategic business decisions and improve operations. We are an interdisciplinary team of Economists, Engineers, and Scientists incubating and building day one solutions using cutting-edge technology, to solve some of the toughest business problems at Amazon.You will work with business leaders, scientists, and economists to translate business and functional requirements into concrete deliverables, including the design, development, testing, and deployment of highly scalable distributed services. You will partner with scientists, economists, and engineers to help invent and implement scalable ML, RL, and econometric models while building tools to help our customers gain and apply insights. This is a unique, high visibility opportunity for someone who wants to have business impact, dive deep into large-scale economic problems, enable measurable actions on the Consumer economy, and work closely with scientists and economists. We are particularly interested in candidates with experience building predictive models and working with distributed systems.As a Senior Applied Scientist, you bring business and industry context to science and technology decisions. You set the standard for scientific excellence and make decisions that affect the way we build and integrate algorithms. Your solutions are exemplary in terms of algorithm design, clarity, model structure, efficiency, and extensibility. You tackle intrinsically hard problems, acquiring expertise as needed. You decompose complex problems into straightforward solutions.
US, WA, Seattle
Do you want to join Alexa AI -- the science team behind Amazon’s intelligence voice assistance system? Do you want to utilize cutting-edge deep-learning and machine learning algorithms to delight millions of Alexa users around the world?If your answers to these questions are “yes”, then come join the Alexa Artificial Intelligence team. We are responsible for the deep learning technology that is central to the automated ranking and arbitration to optimize for end-to-end customer satisfaction.As an Applied Science Manager you will lead the science efforts to develop novel algorithms and modeling techniques to advance the state of the art in spoken language understanding. You will also:· Build a strong and coherent team with particular focus on automated ranking and arbitration, sciences, and innovation.· Serve as a technical lead on demanding and cross-team projects, and effectively collaborating with multiple cross-organizational teams· Apply technical influence on partner teams, increasing their productivity by sharing your deep knowledge.
US, WA, Seattle
Are you interested in delighting customers and are passionate about promoting sustainability? Then Amazon’s Packaging Team is the place for you. We are the Customer Packaging EXperience (CPEX) team and we optimize Amazon’s packaging solutions. To do this across billions of shipments, we are looking for someone with statistics and machine learning skills to design, test, implement and maintain packaging decision mechanisms across Amazon.You'll be responsible for the design, implementation, operation, and support of large-scale, performance-critical data science and machine learning systems which help to choose the optimal packaging types for our products. These technologies utilize every quadrant of data science including statistics, NLP, and CV to make the right decision on millions of products. You will work with scientists, software developers, business intelligence engineers, and product managers on our team as well as partner with packaging automation, concessions, and sustainability teams.
US, WA, Seattle
Have you ever ordered a product on Amazon and wondered how that box with a smile arrived at your doorstep so fast? Wondered where it came from and how much it cost Amazon? If so, the Amazon Global Supply Chain Optimization organization is for you.Watch this video to learn more about our organization, SCOT: are the most customer-centric company on Earth. We need exceptionally talented, bright, and driven people to continue to raise the bar on customer experience.Our objective is to build an experience where you can find and buy anything online and have it delivered fast! We continue to innovate with delivery speed initiatives so that Amazon will continue to own ‘fast’ in the minds of our customers. We are looking for a dynamic, organized self-starter to join as a Research Scientist, who will create state of the art models on Speed initiatives and programs and develop the measurement criteria for success.The AIM (Automated Inventory Management) team in the Supply Chain Optimization Technologies (SCOT) organization is dedicated to answering key strategic questions using quantitative and statistical methods. We develop cutting edge data pipelines, build accurate predictive models, and deploy automated software solutions to provide insights to business leaders at the most senior levels throughout the company. We are looking for a talented, driven, and analytical researcher to help us answer these strategic questions.The AIM Science team leads the way in developing innovative models, algorithms and strategies that will help us gain insights into how our business will grow and what will the drivers of such growth. These predictive models and insights will be based along products and product categories, customer segments, regions and locations, etc.This Research Scientist role will explore and develop innovative quantitative approaches and models, generate features, test hypotheses, design experiments, build predictive models, and work with very large complex data sets in order to explore relationship between business outcomes and key drivers, and then predict the trend on those business outcomes. These predictions and insights will provide a foundation of the highest level of visibility and importance for Amazon's financial and operational planning. The successful candidate will be a problem solver who enjoys diving into data, is excited by difficult modeling challenges, and possesses strong communication skills to effectively interface between technical and business teams, working together with Software Engineers, Product Managers, Business Analysts and other Scientists.Key Responsibilities:· Research, develop and build predictive models for Amazon business metrics with the goal of higher customer satisfactions. Analyze and research features and engineer features that help support predictive models to connect the dots among different functions such as inventory, speed of delivery, and best selections.. Provide insights by analyzing historical data· Constructively critique peer research and mentor junior scientists and engineers.· Create experiments and prototype implementations of new learning algorithms and prediction techniques.· Collaborate with engineering teams to design and implement software solutions for science problems.· Contribute to progress of the Amazon and broader research communities by producing publications.Amazon is an Equal Opportunity Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation.
US, WA, Virtual Location - Washington
Amazon has built a reputation for excellence with recent examples of being named the #1 most trusted company for customers. The Selling Partner Abuse team's mission is to protect the trust of our stores for customers and selling partners. We see ourselves as the steward of customer trust and Amazon brand.We are seeking a Machine Learning Science Manager who will lead a team of Applied Scientists, Research Scientists and Data scientists to research and develop innovative machine learning and science solutions to prevent the bad activities in the Amazon stores.Your responsibilities will include:· Build and develop the core science team for Selling Partner Abuse· Create new projects to drive significant business impact and research advancement· Create project milestone and manage multiple projects end-to-end while quickly adapt to changing priorities and generate innovative solutions in an extremely fast-paced environment· Coach the team and continuously raise the bar on highest standard· Build a strong partnership across different business, engineering and science stakeholders· Manage different Machine Learning projects that cover different science areas such as Ensemble Tree learning models, Clustering, Anomaly Detection, Graph models, NLP models, Semi-supervised Learning models and Reinforcement Learning
AU, VIC, Melbourne
Amazon delights millions of customers around the world. Meet the behind the scenes team that enables our Human Resource and Operations Leaders to make informed decisions. The Amazon PeopleInsight team builds reporting and analytics tools for our teams that fulfill customer promise every day. Whether it is Fulfillment Center team that delivers your Prime order in two days, our Amazon Locker team that lets you pick up your package anytime that is convenient for you, our Prime Now team getting you lunch in under an hour, or one of many more, the PeopleInsight group is there providing people metrics along the employee lifecycle for our global operations businesses. The PeopleInsight team is a collaborative group of Business Analysts, Business Intelligence Engineers, Data Engineers, Data Scientists, Product Managers, and Program Managers dedicated to empowering leaders and enabling action through data and science. We deliver workforce, associate experience, and leadership insights so Amazon leaders can focus their efforts in ways that will engage, retain and grow their associates.We are now recruiting for an exceptional Data Scientist, Worldwide OperationsThe ideal candidate will be:· A Well-Rounded Athlete –Like a true athlete, you understand that we succeed or fail as a team. You are always ready to step up beyond your core responsibilities and go the extra mile for the project and your team. You nimbly overcome barriers to deliver the best products more quickly than expected.· A Perpetual Student – You seek knowledge and insight. You challenge yourself to turn moments into master’s classes. Whether closing a gap, developing a new skill, or staying ahead of your industry, you revel in the joy of learning and growing.· A Skilled Communicator – You excel when interacting with business and technical partners whether you are chatting, sending a written message, or conducting a presentation.· A Trusted Advisor – You work closely with stakeholders to define key business needs and deliver on commitments. You enable effective decision making by retrieving and aggregating data from multiple sources and compiling it into a digestible and actionable format.· An Inventor at Heart – You innovate on behalf of your customer by proactively implementing improvements, enhancements, and customizations. Your customers marvel at your creative solutions to challenges they had not yet identified.· A Fearless Explorer – You are drawn to take on the hardest problems, navigate ambiguity, and battle skepticism. You never settle, even in the face of overwhelming obstacles.Roles and ResponsibilitiesSuccess in this role will include influencing within your team and mentoring peers. The problems you will consider will be difficult to solve and often require a range of data science methodologies combined with subject matter expertise. You will need to be capable of gathering and using complex data set across domains. You will deliver artifacts on medium size projects, define the methodology, and own the analysis. Your findings will affect important business decisions. Solutions are testable and reproducible. You will create documents and share findings in line with scientific best practices for both technical and nontechnical audiences.Amazon is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation.
US, CA, Virtual Location - California
The Amazon Economics Team is hiring Interns in Economics. We are looking for detail-oriented, organized, and responsible individuals who are eager to learn how to work with large and complicated data sets. Some knowledge of econometrics, as well as basic familiarity with Stata or R is necessary, and experience with SQL, UNIX, and Sawtooth would be a plus.These are full-time positions at 40 hours per week, with compensation being awarded on an hourly basis. You will learn how to build data sets and perform applied econometric analysis at Internet speed collaborating with economists, data scientists and MBAʼs. These skills will translate well into writing applied chapters in your dissertation and provide you with work experience that may help you with placement.Roughly 50% of research assistants from previous cohorts have converted to full time data science or economics employment at Amazon. If you are interested, please send your CV to our mailing list at is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation
US, CA, Palo Alto
The Amazon Search team creates powerful, customer-focused search and advertising solutions and technologies. Whenever a customer visits an Amazon site worldwide and types in a query or browses through product categories, A9 Product Search services go to work. We design, develop, and deploy high performance, fault-tolerant distributed search systems used by millions of Amazon customers every day. Our Search team works to maximize the quality and trustworthiness of the search experience for visitors to Amazon websites worldwide.Our mission is to provide customers' trust and confidence in Amazon Search shopping experience. We identify problems that are customer trust busters at Amazon, deliver scalable and responsive solutions to these issues, and build experiences that gain customer trust using advanced machine learning methods. We carefully monitor the trustworthiness of the search results and dive deep when we see an unusual pattern. Most of the models used by our team is semi-supervised or unsupervised using small amount of labeled data.In this role you will leverage your strong statistical background to help build the next generation of our machine learning methods to discover untrustworthy search engagements, unsual patterns, and estimate a probability of risk for each item. 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 machine learning models, graph algorithms, and information retrieval algorithms at scale. Additionally, we are seeking candidates with strong rigor in applied sciences and engineering, creativity, curiosity, and great judgment.
IN, KA, Bengaluru
Do you want to join an innovative team of scientists who use machine learning and statistical techniques to create state-of-the-art solutions for providing better value to Amazon’s customers? Do you want to build and deploy advanced algorithmic systems that help optimize millions of transactions every day? Are you excited by the prospect of analyzing and modeling terabytes of data to solve real world problems? Do you like to own end-to-end business problems/metrics and directly impact the profitability of the company? Do you like to innovate and simplify? If yes, then you may be a great fit to join the Machine Learning and Data Sciences team for India Consumer Businesses.If you have an entrepreneurial spirit, know how to deliver, love to work with data, are deeply technical, highly innovative and long for the opportunity to build solutions to challenging problems that directly impact the company's bottom-line, we want to talk to you.Major responsibilities· · Use machine learning and analytical techniques to create scalable solutions for business problems· · Analyze and extract relevant information from large amounts of Amazon’s historical business data to help automate and optimize key processes· · Design, development, evaluate and deploy innovative and highly scalable models for predictive learning· · Research and implement novel machine learning and statistical approaches· · Work closely with software engineering teams to drive real-time model implementations and new feature creations· · Work closely with business owners and operations staff to optimize various business operations· · Establish scalable, efficient, automated processes for large scale data analyses, model development, model validation and model implementation· · Mentor other scientists and engineers in the use of ML techniques
US, MA, Cambridge
MULTIPLE POSITIONS AVAILABLEEntity: Services LLCJob Title: Applied Scientist IILocation: Cambridge, MAPosition Responsibilities:Designing, developing, evaluating, deploying, and updating data-driven models and analytical solutions for speech and language understanding applications. Developing practical solutions to complex problems. Translating business and functional requirements into concrete deliverables and building quick prototypes or proofs of concept with team members. Solving technical challenges related to speech and language understanding technologies. Mentoring junior engineers and is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation #0000
IN, KA, Bangalore
Amazon is investing heavily in building a world class advertising business and we are responsible for defining and delivering a collection of self-service performance advertising products that drive discovery and sales. Our products are strategically important to our Retail and Marketplace businesses driving long term growth. We deliver billions of ad impressions and millions of clicks daily and are breaking fresh ground to create world-class products. We are highly motivated, collaborative and fun-loving with an entrepreneurial spirit and bias for action. With a broad mandate to experiment and innovate, we are growing at an unprecedented rate with a seemingly endless range of new opportunities.The Moderation and Relevance System (MARS) team, based in Bangalore, is responsible for ensuring that ads are relevant and is of good quality, leading to higher conversion for the sellers and providing a great experience for the customers. We deal with one of the world’s largest product catalog, handle billions of requests a day with plans to grow it by order of magnitude and use automated systems to validate tens of millions of offers submitted by thousands of merchants in multiple countries and languages.In this role, you will build and develop ML models to address content intelligence problems, build advanced algorithms in detecting and generating content. These models will rely on a variety of visual and textual features requiring expertise in both domains. These models need to scale to multiple languages and countries. You will collaborate with engineers and other scientists to build, train and deploy these models. You will propose hypotheses, validate these offline and run A/B tests to validate them online. As part of these activities, you will develop production level code that enables moderation of millions of ads submitted each day.