Classical Machine Learning Amazon.png
Amazon

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 “Amazon.com 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 Amazon.com 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 Amazon.com 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.

Reconstruction

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.
Amazon

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 Amazon.com that much more useful for customers.

Work with us

See More Jobs
AU, NSW, Sydney
DESCRIPTIONAmazon Web Services is the largest consumer cloud offering in the world, powering cutting-edge science, rapidly growing startups, and industry leaders. Our team here in Sydney is developing novel products focused on helping our customers achieve higher availability standards for the applications and workloads they build on AWS. We are building solutions that aim to learn from past incidents data to improve automatic behaviours and help investigate and resolve incidents faster in real time. We are working on complex problems including automatic incident investigation and root-causing, as well as other ideas for assisting our customers. This is an exciting opportunity if you love solving challenging problems and contributing to high impact customer features involving real-life data and Machine Learning methods, while working within a development team to see the results deployed in production, and being part of the large global community of Amazon scientists.We are looking for an Applied Scientist who is enthusiastic about diving deep into the data to analyse and uncover relevant patterns, and is excited to go beyond the standard toolbox when needed, to research and create novel methods for producing the best possible solutions to the problems our customers are facing. The methodology and algorithm development are highly coupled here with deep data analysis and understanding of the real-life data.Are you interested in solving complex real-life challenges which are not easily translatable to textbook examples, where the data is rarely clean or readily available, and its amount varies from immense to lacking? We need a practitioner who loves designing a new model or an algorithm, and also working with the rest of the development team to implement it efficiently in code and deploy to production.While working on these problems, you will help to grow the Machine Learning and Data Science team and the ML community in Sydney, collaborate with the global Amazon research community, and publish research results. You will engage with other teams across Amazon who are working on related problems, and communicate with product management and stakeholders to provide clear answers for business questions, present the findings and new proposals.In this role you will build something brand new that will positively affect every AWS customer and in turn their customers, while being at the forefront of using and developing novel Machine Learning techniques to deliver new functionality. Join us, and help us revolutionise how people approach availability!Responsibilities Summary· Design and implement predictive models based on past data, using either existing methods or developing new algorithms and tools when needed.· Collect and analyze data, develop hypotheses, design experiments, collaborate with engineering team to implement live tests and evaluate performance.· Answer complex business questions by using appropriate statistical techniques on available data or designing and running experiments to gather data.· Design, develop, and operate critical processes that are central to the delivery of accurate results (e.g. data pipelines).· Code in a high-level language (e.g. Python, Java) and work with software engineers to produce high-quality software in production, based on research models.· Communicate findings to managers and engineers, often through succinct written summaries of findings and code samples, as well as through verbal explanation.· Publish research results.
US, CA, San Francisco
The new Amazon Scholars program has broadened opportunities for academics to join Amazon in a flexible capacity, in particular part-time arrangements and sabbaticals.The Amazon Scholars program is designed for academics from universities around the globe who want to apply research methods in practice and help us solve hard technical challenges without leaving their academic institutions. We believe that Amazon is a unique place to measure the impact of new scientific ideas, given our scale and our ownership of both an information infrastructure and physical infrastructure. You will have a chance to have a ground-up impact on our systems, our business, and most importantly, our customers, through your expertise.Applications are accepted from academic experts in research areas including, but not limited to, the following: Artificial Intelligence, Avionics, Computer Vision, Data Science, Economics, Machine Learning, Optimization, Natural Language Processing, and Robotics.As an Amazon Scholar, your responsibilities may include 1) advising business leaders on strategic plans, 2) diving deep to solve a specific technical problem in an organization’s roadmap, and 3) advising junior researchers on methods.We’re working on the future here at Amazon. Come join us.
US, CA, San Francisco
Are you interested in building, leading and driving the machine learning technical vision, strategy and implementation of Personalized ML products for Customers? Build and lead an innovative team of scientists and engineers who will invent and apply advanced machine learning science and techniques to create the best customer experience on the earth.The ideal candidate will be an expert in the areas of reccomender systems, data science, machine learning and statistics. You will have hands-on experience leading multiple simultaneous product development initiatives as well as be able to balance technical leadership with strong business judgment to make the right decisions about technology, models and methodologies choices. You will strive for simplicity, and demonstrate significant creativity and high judgment backed by statistical proof.
US, WA, Bellevue
The Alexa Experience organization owns a number of programs and domains which drive high customer engagement and feature discovery, maintains customer trust, understands customers’ feedback, develops new features that provide utility value for customers with special needs, and develops locally relevant experiences. Our team develop and train natural language understanding models to improve these programs and to create new delighting Alexa experiences for customers.We’re looking for a passionate scientist to help design, develop and release to production NLU models that provide the best-possible experience for our customers. You will leverage Amazon’s heterogeneous data sources and large-scale computing resources to develop new features and predict key user behaviors, both offline and in real time.What You Will Do· Be the subject matter expert in NLU model authoring process, guiding multiple product teams toward delivering best customer experiences on Alexa.· Contribute to the design of new features by extending existing annotation schemas to cover new areas· Develop and release to production Alexa NLU models· Measure and analyze accuracy of models; develop solutions to improve performances· Contribute to the research vision of the team, develop and execute on a roadmap that addresses the technical challenges our product teams face· Raise the bar for research quality and impact. Stay up to date on research results both within and complementary to the Alexa technology space and helping your team and its partner to utilize tried patterns and state of the art results to work smarter and faster.· Propose new modeling solutions and run experiments to prove their efficacy· Deliver reports and/or models to influence technical directions.· Design processes and tools to simplify Alexa feature expansion for your team and partners.· Perform data analysis, A/B experimentation and modeling with large data sets to develop insights that increase device usage and customer experience· Collaborate with peers in data analysis and feature design efforts
US, CA, San Francisco
We are looking for a passionate and talented Applied Scientist who will collaborate with other scientists and engineers to leverage machine learning methods and algorithms for the modeling and analysis of data.You'll design and run experiments, research new algorithms, and find new ways of optimizing customer experience. Besides theoretical analysis and innovation, you will work closely with talented engineers to put your algorithms and models into practice. Your work will directly impact the trust customers place in Amazon.You are the ideal candidate if you are passionate about delivering experiences that delight customers and creating solutions that are robust. You thrive in ambiguous environments that require to find solutions to problems that have not been solved before. You leverage your exceptional technical expertise, a sound understanding of the fundamentals of Computer Science, and practical experience to creating reliable, scalable, and high performance products. Your strong communication skills enable you to work effectively with both business and technical partners. You have hands-on experience making the right decisions about technology, models and methodology choices. You strive for simplicity, and demonstrate significant creativity and high judgment backed by statistical proof.Responsibilities include:· Design, implement, test, deploy and maintain innovative data and machine learning solutions to further the customer experience.· Create experiments and prototype implementations of new learning algorithms and prediction techniques· Collaborate with scientists, engineers, product managers and business stakeholders to design and implement software solutions· Use machine learning best practices to ensure a high standard of quality for all of the team deliverables
US, WA, Seattle
Are you interested in building and driving the technical vision, strategy, and implementation for AWS Hardware? AWS Hardware is hiring a Senior Data Scientist to help lead the definition and prioritization of our customer focused technologies and services. AWS Hardware is responsible for designing, qualifying, and maintaining server solutions for AWS and its customers as well as developing new cloud focused hardware solutions. You will be a senior technical leader in the existing Data Sciences and Analytics Team, build, and drive the data science and analytics needed for our product development and operations.As a Senior Data Scientist, you are considered a data science leader on your team. You take the lead on large projects and drive solutions to complex or ambiguous problems. You design solutions that help us take advantage of business opportunities (new and existing). You may own the delivery of data science solutions for an entire business application. You are a key influencer of your team’s business strategy are able to influence the strategies of related teams. You work efficiently and routinely deliver the right things with limited guidance. Last, you are externally aware of the state-of-the-art in the data science field and are constantly focused on advancing the state-of-the-art for improving Amazon’s products and services.The ideal candidate will be able to solve complex or critical problems using data science principles that are highly accurate, stable, secure, scalable, and cost-effective. You will have hands-on experience leading multiple simultaneous product development initiatives as well as be able to balance technical leadership with strong business judgment to make the right decisions about technology, models and methodologies choices.You will strive for simplicity, and demonstrate significant creativity and high judgment backed by statistical proof. As a Senior Data Scientist, you will be responsible for developing KPIs, programs, tests and research in order to drive business solutions using computational techniques, and statistics to support strategic and tactical decision-making. The ideal candidate will have significant experience performing large-scale data analysis and reporting and will be responsible for developing, improving and simplifying our models. This person has the opportunity to leverage extensive datasets and provide meaningful insights to improve the customer experience as well as optimize business decisions for Amazon.This position requires a solution-oriented candidate with a combination of deep business acumen, expert knowledge of statistics and algorithm development, and an analytical mindset. The candidate must have the ability to work with diverse stakeholder groups to solve business problems and provide data solutions that are organized and simple to understand. This is a highly visible position that will interact at all levels of the business. Become part of this unique opportunity to make history defining the evolution of cloud computing.
LU, Luxembourg
Amazon is looking for a Data Scientist to provide statistical and analytical insights from the vast array of services powering one of the world’s largest Reliability, Maintenance Engineering (RME) platforms. As part of the Automation Engineering team, we support the business in all aspects of the traditional automation pyramid, and we provide our internal users information on all aspects of the status of control systems within Amazon’s EU Fulfillment Network. At a basic level, these systems link our low-level automation systems with the cloud and we work at the cutting edge of all aspects of the automation pyramid, from device level to the enterprise level. The more complex systems are leveraging Machine Learning and Big Data to drive predictive actions, preventing downtime or defects in the material handling equipment and improving the Overall Equipment Efficiency of the installations.This role requires an individual with excellent statistical and analytical abilities, professional experience applying data science methodologies and data engineering practices as well as outstanding business acumen and ability to work with various teams across Amazon. The successful candidate will be a self-starter comfortable with ambiguity, with strong attention to detail, an ability to work in a fast-paced and ever-changing environment, and driven by a desire to innovate in this space.Responsibilities:You are a significant and autonomous contributor. Your work is consistently of high quality. You are able to use a range of data science methodologies to conduct analysis for cases when the solution approach is unclear that relate to a portion of a business or business process. You apply a breadth of experience from practical application of data science techniques and tools to solve difficult business problems. Your work is focused on team-level goals, medium size projects, and small subsets of larger goals. You are able to ramp up quickly on new areas where colleagues identify established solutions. You are expected to consistently demonstrate a combination of the following:· You independently own and solve difficult business problems. These may be well-defined problems where the solution has not yet been outlined or approach to solve is unclear.· You deliver artifacts on medium size projects that affect important business decisions. You define the methodology and own the analysis.· You are able to gather and use complex data set across domains. You proactively gather data when it is not readily available.· You skilfully employ a range of data science methods, tools, and best practices. You are able to justify your approach.· You write clear and factually correct documents explaining technical concept to non-technical audience.· You have good working relationships with team-mates and peers working on related areas. You recognize discordant views and take part in constructive dialogue to resolve them.· You confidently train new team-mates about your customers, how your team’s solutions work, and how those solutions are reflected in the data.
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. We are looking for a highly motivated, top notch applied scientist to build machine learning models at scale to enforce our policy guidelines. A successful candidate will have demonstrated experience in at least some of the following areas: NLP, Image Recognition and Classification, Video Recognition and Classification, Generative Models, Reinforcement Learning, Active Learning, Weak SupervisionYour areas of responsibility include:· · Designing and implementing new features and machine learned models, including the application of state-of-art deep learning to solve ad policy enforcement and creative intelligence, including NLP, deep image and video models, generative models · Perform analysis of data and metrics relevant to ad content generation and policing · Gathering ad policy related requirements from business owners, other tech teams, as well as by analyzing customer feedback and translate them into modeling problems · Integrate and productize ML models with overall engineering infrastructure to be made available at scaleAd Quality protects the customer experience and is a critical component of our business success. One of the earliest teams to be established in Amazon Bangalore, Ad Quality has both Operations and Development teams in Bangalore supporting multiple ad programs in markets around the world.sspajobs
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. We are looking for a highly motivated, top notch applied scientist to build machine learning models at scale to enforce our policy guidelines. A successful candidate will have demonstrated experience in at least some of the following areas: NLP, Image Recognition and Classification, Video Recognition and Classification, Generative Models, Reinforcement Learning, Active Learning, Weak SupervisionYour areas of responsibility include:· · Designing and implementing new features and machine learned models, including the application of state-of-art deep learning to solve ad policy enforcement and creative intelligence, including NLP, deep image and video models, generative models · Perform analysis of data and metrics relevant to ad content generation and policing · Gathering ad policy related requirements from business owners, other tech teams, as well as by analyzing customer feedback and translate them into modeling problems · Integrate and productize ML models with overall engineering infrastructure to be made available at scaleAd Quality protects the customer experience and is a critical component of our business success. One of the earliest teams to be established in Amazon Bangalore, Ad Quality has both Operations and Development teams in Bangalore supporting multiple ad programs in markets around the world.sspajobs
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. We are looking for a highly motivated, top notch applied scientist to build machine learning models at scale to enforce our policy guidelines. A successful candidate will have demonstrated experience in at least some of the following areas: NLP, Image Recognition and Classification, Video Recognition and Classification, Generative Models, Reinforcement Learning, Active Learning, Weak SupervisionYour areas of responsibility include:· · Designing and implementing new features and machine learned models, including the application of state-of-art deep learning to solve ad policy enforcement and creative intelligence, including NLP, deep image and video models, generative models · Perform analysis of data and metrics relevant to ad content generation and policing · Gathering ad policy related requirements from business owners, other tech teams, as well as by analyzing customer feedback and translate them into modeling problems · Integrate and productize ML models with overall engineering infrastructure to be made available at scaleAd Quality protects the customer experience and is a critical component of our business success. One of the earliest teams to be established in Amazon Bangalore, Ad Quality has both Operations and Development teams in Bangalore supporting multiple ad programs in markets around the world.sspajobs
US, WA, Seattle
Amazon is looking for a passionate, talented, and inventive Senior Scientist with a strong machine learning background to help build industry-leading Speech and Language technology. Our mission is to push the envelope in Automatic Speech Recognition (ASR), Natural Language Understanding (NLU), and Audio Signal Processing, in order to provide the best-possible experience for our customers.As a Senior Scientist, you will work with talented peers to develop novel algorithms and modeling techniques to advance the state of the art in spoken language understanding. Your work will directly impact our customers in the form of products and services that make use of speech and language technology. You will leverage Amazon’s heterogeneous data sources and large-scale computing resources to accelerate advances in spoken language understanding.We are hiring in all areas of spoken language understanding: ASR, NLU, text-to-speech (TTS), and Dialog Management.
US, TN, Nashville
Workforce Staffing (WFS), a division of Amazon’s Worldwide Operations Human Resources, manages Amazon’s Tier 1 talent supply chain. We attract, hire, and onboard the associates who, by fulfilling orders at the frontlines of the company, make Amazon a global leader in delivery and logistics. The mission of the Candidate Experience Research Team (CER) is to drive positive change to Workforce Staffing’s hiring and onboarding experience. We conduct research and analyses on opportunities and successes for the business, and develop products that amplify the voice of our candidates.We are seeking a manager, data science, with a heavy focus on quantitative data analysis and evaluation, and a deep focus on understanding labor markets. You will be responsible for building a new team from the ground up, develop roadmaps, and drive business impact through your research at global scale.The ideal candidate should be well versed in quantitative methods, including classical statistics and machine learning approaches. Competitive candidates will be very comfortable with at least one computational language (e.g., R, Python). Candidates should be comfortable combing through computational models, machine learning algorithms, and analyzing their output.Candidates should have demonstrated experience leading data science projects related to labor market research and analysis, including research on wage sensitivity and elasticity, workforce relevance, and other factors.A customer-obsessed, relentless curiosity is a must, as is commitment to the highest standards of methodological rigor that a given study allows. This role provides opportunity for significant exposure to Amazon’s culture, leadership, and global businesses, and furthermore provides significant opportunity to influence how Workforce Staffing matches talent to business demand.This role will be located in Nashville.If you’re hungry to engage and empower Amazon Associates your expertise in mixed methods research, let's talk.
US, WA, Seattle
Workforce Staffing (WFS), a division of Amazon’s Worldwide Operations Human Resources, manages Amazon’s Tier 1 talent supply chain. We attract, hire, and onboard the associates who, by fulfilling orders at the frontlines of the company, make Amazon a global leader in delivery and logistics. The mission of the Candidate Experience Research Team (CER) is to drive positive change to Workforce Staffing’s hiring and onboarding experience. We conduct research and analyses on opportunities and successes for the business, and develop products that amplify the voice of our candidates.We are seeking a research scientist with deep expertise in diversity research, with a heavy focus on quantitative data analysis and evaluation.The ideal candidate should be well versed in quantitative methods, including classical statistics and machine learning approaches. Competitive candidates will be very comfortable with at least one computational language (e.g., R, Python). Candidates should be comfortable combing through computational models, machine learning algorithms, and analyzing their output.Candidates should have demonstrated experience leading research projects related to diversity, including program evaluation. The ideal candidate will have demonstrated experience auditing programs, and advocating for and implementing solutions.A customer-obsessed, relentless curiosity is a must, as is commitment to the highest standards of methodological rigor that a given study allows. This role provides opportunity for significant exposure to Amazon’s culture, leadership, and global businesses, and furthermore provides significant opportunity to influence how Workforce Staffing matches talent to business demand.If you’re hungry to engage and empower Amazon Associates your expertise in mixed methods research, let's talk.
US, WA, Seattle
Want to transform the way people enjoy music, video, and radio? Come join Alexa Entertainment, the team that made Amazon Music, Spotify, Hulu, Netflix, Pandora, and more available to Alexa customers. We are at the epicenter of the future of entertainment, innovating the way our customers interact with content at home and on the go.Alexa Entertainment is looking for a Senior Applied Scientist to lead a new team of talented and passionate scientists to intelligently arbitrate among competing possible actions to take on behalf of customers, as part of our wider spoken language understanding efforts. As a Senior Applied Scientist, you will lead the design, development, and evaluation of models and ML (machine learning) technology to enable a magical entertainment experience for Alexa customers. You will help lay the foundation to move from directed interactions to learned behaviors that enable Alexa to proactively take action on behalf of the customer. And, you will have the satisfaction of working on a product your friends and family can relate to, and want to use every day. Like the world of smart phones less than 10 years ago, this is a rare opportunity to have a giant impact on the way people live.
US, WA, Seattle
Are you passionate about building data-driven solutions to drive the profitability of the business? Are you excited about solving complex real world problems? Do you have proven analytical capabilities, exceptional communication, project management skills, and the ability to multi-task and thrive in a fast-paced environment? We need your skills and experience to help us grow Amazon Lending.Amazon Lending is an exciting program that provides businesses with click-to-cash access to funds that they need to grow their business with Amazon. We take pride in building scalable automated systems that make data-driven decisions to automate the entire lending process on a global scale.
US, WA, Seattle
Are you passionate about building data-driven solutions to drive the profitability of the business? Are you excited about solving complex real world problems? Do you have proven analytical capabilities, exceptional communication, project management skills, and the ability to multi-task and thrive in a fast-paced environment? We need your skills and experience to help us grow Amazon Lending.Amazon Lending is an exciting program that provides businesses with click-to-cash access to funds that they need to grow their business with Amazon. We take pride in building scalable automated systems that make data-driven decisions to automate the entire lending process on a global scale.
US, CA, Santa Clara
The Household Organization (HHO) team builds the conversations you and your family have with Alexa that help you get and stay organized. Millions of customers use our products to manage their time, get things done, and access and manage their personal information with their voice.We are making Alexa smarter and more personalized and are proud to do this in a privacy first way. We are looking to hire a machine learning scientist with expertise in NLP, differential privacy and data generation.
US, WA, Seattle
The IMDb TV team is shaping the future of digital video entertainment. Our mission is to build earth’s most customer centric ad supported premium free video service and make it trivially easy for hundreds of millions of customers to enjoy. Our team is re-inventing how to find content and building a new video destination.We need to make agile decisions based on what content creates the most value for our customers and pursue the most efficient content acquisition strategies for our desired outcomes. We are seeking an innovative Data Scientist to predict and measure the benefit of different IMDb TV titles and their impact on customer engagement as well as advertiser value. This position will be responsible for designing and building the suite of models that will predict content performance as well as the interface used by our content acquisition team to assess content for use in real time deal evaluations.This role requires a team member with strong quantitative modeling skills and the ability to apply statistical/machine learning methods to large amounts of customer and title level data. The candidate should have strong communication skills, be able to work closely with stakeholders and translate data-driven findings into actionable insights. The successful candidate will be a self-starter comfortable with ambiguity, with strong attention to detail and an ability to work in a fast-paced and ever-changing environment. This person will also be proficient in communicating their recommendations directly to advertising and video leadership.This position will be part of the Prime Video Content Research team, which includes a diverse scientific team of computer scientists and economists as well as other data scientists who build statistical models using world-class data systems and partner directly with the business to implement the solutions. We use detailed customer behavioral data (e.g. streaming history) and detailed information about content (e.g. IMDb-sourced characteristics) to predict and understand what customers like to watch. As a Data Scientist at Amazon, you will have the opportunity to work on one of the world's largest consumer data sets and influence the long term evolution of our analytics capability.This is very much Day 1 for IMDb TV. You would be joining an entrepreneurial and pioneering team working to reinvent ad supported television.
US, NY, New York
About Amazon Advertising:Amazon is building a world class advertising business and defining and delivering a collection of self-service performance advertising products that drive discovery and sales of merchandise. 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.About our team:Our team's mandate is to accelerate advertising in the Books category as it requires a differentiated customer and advertiser experience. We own end-to-end the advertising experience including placements, ad relevance, creative, ad serving, advertiser experience, and marketing. We are looking for entrepreneurial, innovative individuals who thrive on solving tough problems. We are investing in a deep science and technical team to pursue a transformation opportunity.About this role:We are seeking a talented, energetic, entrepreneurial, and self-driven Senior Data Scientist to join our team. Our team works on complex science, engineering, optimization, econometric, and user-experience problems in order to deliver relevant product ads on Amazon search and detail pages world-wide. Leveraging Amazon's massive data repository, our data scientists develop experiments, insights and optimizations that enable the monetization of Amazon online and mobile search properties while enhancing the experience of Amazon shoppers.In this role, you will:· Solve new and challenging problems using statistical, Machine Learning, or optimization approaches to create measurable customer facing impact.· Translate prototype models to production quality, scaling to millions of customers and across different languages.· Design experiments leveraging your models to measure effectiveness.· Work closely with product, business and engineering teams to execute on your project plans.· Communicate verbally and in writing to senior leaders with various levels of technical knowledge, educating them about your approach, as well as sharing insights and recommendations.· Drive best practices on the team; mentor and guide junior members to achieve their career growth potential.
US, MA, Cambridge
Alexa is the groundbreaking voice service that powers Echo and other devices designed around your voice. Our team is creating the science and technology behind Alexa. We’re working hard, having fun, and making history. Come join our team! You will have an enormous opportunity to impact the customer experience, design, architecture, and implementation of a cutting edge product used every day by people you know.We’re looking for a passionate, talented, and inventive scientist to help build industry-leading conversational technologies that customers love. Our team's mission is the enable Alexa to understand sounds and vocalization beyond speech. As a Research Scientist, you will work with talented peers to develop novel algorithms and modeling techniques to advance the state of the art in speech and audio processing. Your work will directly impact our customers in the form of novel products and services that make use of speech and language technology. You will leverage Amazon’s heterogeneous data sources and large-scale computing resources to accelerate advances in spoken language understanding. You will mentor junior scientists, create and drive new initiatives.