Card-Imbens 16x9.jpg
David Card (left), an Amazon Scholar, a professor of economics at the University of California, Berkeley, and the outgoing president of the AEA, and Guido Imbens (right), an academic research consultant at Amazon and a professor at the Stanford Graduate School of Business.

A conversation with economics Nobelists

Amazon Scholar David Card and academic research consultant Guido Imbens on the past and future of empirical economics.

The annual meeting of the American Economic Association (AEA) took place Jan. 7 - 9, and as it approached, Amazon Science had the chance to interview two of the three recipients of the 2021 Nobel Prize in economics — who also happen to be Amazon-affiliated economists.

David Card, an Amazon Scholar, a professor of economics at the University of California, Berkeley, and the outgoing president of the AEA, won half the prize “for his empirical contributions to labor economics”.

Guido Imbens, an academic research consultant at Amazon and a professor at the Stanford Graduate School of Business, shared the other half of the prize with MIT’s Josh Angrist for “methodological contributions to the analysis of causal relationships”.

Amazon Science: The empirical approach to economics has been recognized by the Nobel Prize committee several times in the last few years, but it wasn't always as popular as it is today. I'm curious how you both first became interested in empirical approaches to economics.

David Card: The heroes of economics for many, many decades were the theorists, and in the postwar era especially, there was a recognition that economic modeling was underdeveloped — the math was underdeveloped — and there was a need to formalize things and understand better what the models really delivered.

People started to realize that we had the data to better look at real labor market phenomena and possibly make economics something different than just a kind of a branch of philosophy.
David Card

That need really proceeded through the ’60s, and Arrow and Debreu were these famous mathematical economists who developed some very elegant theoretical models of how the market works in an idealized economy.

What happened in my time was people started to realize that we had the data to better look at real labor market phenomena and possibly make economics something different than just a kind of a branch of philosophy. Arrow-Debreu is basically mathematical philosophy.

Guido Imbens: I came from a very different tradition. I grew up in the Netherlands, and there was a strong tradition of econometrics started by people like Tinbergen. Tinbergen had been very broad — he did econometrics, but he also did empirical work and was very heavily involved in policy analysis. But over time, the program he had started was becoming much more focused on technical econometrics.

So as an undergraduate, we didn't really do any empirical work. We really just did a lot of mathematical statistics and some operations research and some economic theory. My thesis was a theoretical econometrics study.

When I presented that at Harvard, Josh Angrist wasn't really all that impressed with it, and he actually opposed the department hiring me there because he thought the paper was boring. And he was probably right! But luckily, the more senior people there at the time thought I was at least somewhat promising. And so I got hired at Harvard. But then it was really Josh and Larry Katz, one of the labor economists there, who got me interested in going to the labor seminar and got me exposed to the modern empirical work.

The context Josh and I started talking in really was this paper that I think came up in all three of the Nobel lectures, this paper by Ed Leamer, “Let's Take the Con Out of Econometrics”, where Leamer says, “Hardly anyone takes data analysis seriously. Or perhaps more accurately, hardly anyone takes anyone else’s data analysis seriously.”

And I think Leamer was right: people did these very elaborate things, and it was all showing off complicated technical things, but it wasn't really very credible. In fact, Leamer presented a lecture based on that work at Harvard. And I remember Josh getting up at some point and saying, “Well, you talk about all this old stuff, but look at the work Card does. Look at the work Krueger does. Look at the work I do. It's very different.”

And that felt right to me. It felt that the work was qualitatively very different from the work that Ed Leamer was describing and that he was complaining about.

AS: So that's when you first became aware of Professor Card’s work. Professor Card, when did you first become aware of Professor Imbens’s work?

Card: One of his early papers was pretty interesting. He was trying to combine data from micro survey evidence with benchmark numbers that you would get from a population, and it's actually a version of a kind of a problem that arises at Amazon all the time, which is, we've got noisy estimates of something, and we've got probably reliable estimates of some other aggregates, and there's often ways to try and combine those. I saw that and I thought that was very interesting.

Then there’s the problem that Josh and Guido worked on that was most impactful and that was cited by the Nobel Prize committee. I had worked on an experiment, a real experiment [as opposed to a natural experiment], in welfare analysis in Canada, and it was providing an economic incentive to try and get single mothers off of welfare and into work. And we noticed that the group of mothers who complied or followed on with the experiment was reasonable size, but it wasn't 100%.

We did some analysis of it trying to characterize them. Around the same time, I became aware of Imbens’s and Angrist’s paper, which basically formalized that a lot better and described what exactly was going on with this group. That framework just instantly took off, and everyone within a few years was thinking about problems that way.

This morning I was talking to another Amazon person about a problem. It was a difference analysis. I was saying we should try and characterize the compliers for this difference intervention. So it's exactly this problem.

The Nobel committee’s press release for Card, Imbens, and Angrist’s prize announcement emphasizes their use of natural experiments, which it defines as “situations in which chance events or policy changes result in groups of people being treated differently, in a way that resembles clinical trials in medicine.” A seminal instance of this was Card’s 1993 paper with his Princeton colleague Alan Krueger, which compared fast-food restaurants in two demographically similar communities on either side of the New Jersey-Pennsylvania border, one of which had recently seen a minimum-wage hike and one of which hadn’t.

AS: In the early days, there was skepticism about the empirical approach to economics. So every time you selected a new research project, you weren't just trying to answer an economics problem; you were also, in a sense, establishing the credibility of the approach. How did you select problems then? Was there a structure that you recognized as possibly lending itself to natural experiment?

Card: I think that the natural-experiment thing — there was really a brief period where that was novel, to tell you the truth. Maybe 1989 to 1992 or 3. I did this paper on the Mariel boatlift, which was cited by the committee. But to tell you the truth, that was a very modest paper. I never presented it anywhere, and it's in a very modest journal. So I never thought of that paper as going anywhere [laughs].

What happened was, it became more and more well understood that in order to make a claim of causality even from a natural-experiment setting, you had to have a fair amount of information from before the experiment took place to validate or verify that the group that you were calling the treatment group and the group that you were calling the control group actually were behaving the same.

That was a weakness of the project that Alan Krueger and I did. We had restaurants in New Jersey and Pennsylvania. We knew the minimum wage was going to increase — or we thought we knew that; it wasn't entirely clear at the time — but we surveyed the restaurants before, and then the minimum wage went up, and we surveyed them after, and that was good.

But we didn't really have multiple surveys from before to show that in the absence of the minimum wage, New Jersey and Pennsylvania restaurants had tracked each other for a long time. And these days, that's better understood. At Amazon for instance, people are doing intervention analyses of this type. They would normally look at what they call pre-trend analysis, make sure that the treatment group and the control group are trending the same beforehand.

I think there are 1,000 questions in economics that have been open forever. Sometimes new datasets come along. That's been happening a lot in labor economics: huge administrative datasets have become available, richer and richer, and now we're getting datasets that are created by these tech firms. So my usual thing is, I think, that's a dataset that maybe we can answer this old question on. That’s more my approach.

That's why being at Amazon has been great .... A lot of people have substantive questions they're trying to analyze with data, and they're kind of stuck in places, so there's a need for new methodologies.
Guido Imbens

Imbens: I come from a slightly different perspective. Most of my work has come from listening to people like David and Josh and seeing what type of problems they're working on, what type of methods they're using, and seeing if there's something to be added there — if there’s some way of improving the methods or places where maybe they're stuck, but listening to the people actually doing the empirical work rather than starting with the substantive questions.

That's why being at Amazon has been great, from my perspective. A lot of people have substantive questions they're trying to analyze with data, and they're kind of stuck in places, so there's a need for new methodologies. It's been a very fertile environment for me to come up with new research.

AS: Methodologically, what are some of the outstanding questions that interest you both?

Imbens: Well, one of the things is experimental design in complex environments. A lot of the experimental designs we’re using at the moment still come fairly directly from biomedical settings. We have a population, we randomize them into a treatment group and a control group, and then we compare outcomes for the two groups.

But in a lot of the settings we’re interested in at Amazon, there are very complex interactions between the units and their experiences, and dealing with that is very challenging. There are lots of special cases where we know somewhat what to do, but there are lots of cases where we don't know exactly what to do, and we need to do more complex experiments to get the answers to the questions we're interested in.

Double randomization — original color scheme.jpeg
An example of what Imbens calls “experimental design in complex environments”. In this illustration, each of five viewers is shown promotions for eight different Prime Video shows. Some of those promotions contain extra information, indicated in the image by star ratings (the “treatment”). This design helps determine whether the treatment affects viewing habits (the viewer experiment) but also helps identify spillover effects, in which participation in the viewer experiment influences the viewer’s behavior in other contexts.

The second thing is, we do a lot of these experiments, but often the experiments are relatively small. They’re small in duration, and they’re small in size relative to the overall population. You know, it goes back to the paper we mentioned before, combining this observational-study data with experimental data. That raises a lot of interesting methodological challenges that I spend a lot of time thinking about these days.

AS: I wondered if in the same way that in that early paper you were looking at survey data and population data, there's a way that natural experiments and economic field experiments can reinforce each other or give you a more reliable signal than you can get from either alone.

Card: There's one thing that people do; I've done a few of these myself. It's called meta analysis. It's a technique where you take results from different studies and try and put them into a statistical model. In a way it's comparable to work Guido has done at Amazon, where you take a series of actual experiments, A/B experiments done in Weblab, and basically combine them and say, “Okay, these aren't exactly the same products and the same conditions, but there's enough comparability that maybe I can build a model and use the information from the whole set to help inform what we're learning from any given one.”

And you can do that in studies in economics. For example, I’ve done one on training programs. There are many of these training programs. Each of them — exactly as Guido was saying — is often quite small. And there are weird conditions: sometimes it's only young males or young females that are in the experiment, or they don't have very long follow-up, or sometimes the labor market is really strong, and other times it's really weak. So you can try and build a model of the outcome you get from any given study and then try and see if there are any systematic patterns there.

Imbens: We do all these experiments, but often we kind of do them once, and then we put them aside. There's a lot of information over the years built up in all these experiments we've done, and finding more of these meta-analysis-type ways of combining them and exploiting all the information we have collected there — I think it's a very promising way to go.

AS: How can empirical methods complement theoretical approaches — model building of the kind that, in some sense, the early empirical research was reacting against?

Card: Normally, if you're building a model, there are a few key parameters, like you need to get some kind of an elasticity of what a customer will do if faced with a higher price or if offered a shorter, faster delivery speed versus slower delivery speed. And if you have those elasticities, then you can start building up a model.

If you have even a fairly complicated dynamic model, normally there's a relatively small number of these parameters, and the value of the model is to take this set of parameters and try and tell a bit richer story — not just how the customer responds to an offer of a faster delivery today but how that affects their future purchases and whether they come back and buy other products or whatever. But you need credible estimates of those elasticities. It's not helpful to build a model and then just pull numbers out of the air [laughs]. And that's why A/B experiments are so important at Amazon.

AS: I asked about outstanding methodological questions that you're interested in, but how about economic questions more broadly that you think could really benefit from an empirical approach?

Card: In my field [labor economics], we've begun to realize that different firms are setting different wages for the same kinds of workers. And we're starting to think about two issues related to that. One is, how do workers choose between jobs? Do they know about all the jobs out there? Do they just find out about some of the jobs? We're trying to figure out exactly why it's okay in the labor market for there to be multiple wages for a certain class of workers. Why don't all the workers immediately try to go to one job? This seems to be a very important phenomenon.

And on the other side of that, how do employers think about it? What are the benefits to employers of a higher wage or lower wage? Is it just the recruiting, or is it retention, or is it productivity? Is it longer-term goals? That's front and center in the research that I do outside of Amazon.

AS: I was curious if there were any cases where a problem presented itself, and at first you didn't think there was any way to get an empirical handle on it, and then you figured out that there was.

We're supposed to be social scientists who are trying to see what people are doing and the problems they confront and trying to analyze them. ... That's different than this old-fashioned Adam Smith view of the economy as a perfectly functioning tool that we're just supposed to admire.
David Card

Card: I saw a really interesting paper that was done by a PhD student who was visiting my center at Berkeley. In European football, there are a lot of non-white players, and fan racism is pretty pervasive. This guy noticed that during COVID, they played a lot of games with no fans. So he was able to compare the performance of the non-white and white players in the pre-COVID era and the COVID era, with and without fans, and showed that the non-white players did a little bit better. That's the kind of question where you’re saying, How are we ever going to study that? But if you're thinking and looking around, there's always some angle that might be useful.

Imbens: That's a very clever idea. I agree with David. If you just pay attention, there are a lot of things happening that allow you to answer important questions. Maybe fan insults in sports itself isn't that big a deal, but clearly, racism in the labor market and having people treated differently is a big problem. And here you get a very clear handle on an aspect of it. And once you show it's a problem there, it's very likely that it shows up in arguably substantively much more important settings where it's really hard to study.

In the Netherlands for a long time, they had a limit on the number of students who could go to medical school. And it wasn't decided by the medical schools themselves; they couldn't choose whom to admit. It was partly based on a lottery. At some point, someone used that to figure out how much access to medical school is actually worth. So essentially, you have two people who are both qualified to go to medical school; one gets lucky in the lottery; one doesn't. And it turns out you're giving the person who wins the lottery basically a lot of money. Obviously, in many professions we can't just randomly assign people to different types of jobs. But here you get a handle on the value of rationing that type of education.

Card: I think that's really important. You know, we're supposed to be social scientists who are trying to see what people are doing and the problems they confront and trying to analyze them. In a way, that's different than this sort of old-fashioned Adam Smith view of the economy as a perfectly functioning tool that we're just supposed to admire. That is a difference, I think.

Research areas

Related content

US, NY, New York
MULTIPLE POSITIONS AVAILABLE Employer: AMAZON.COM SERVICES LLC Offered Position: Applied Scientist II Job Location: New York, New York Job Number: AMZ10564066 Position 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 scientists. 40 hours / week, 8:00am-5:00pm, Salary Range: $172,400/year to $223,400/year. Amazon is a total compensation company. Dependent on the position offered, equity, sign-on payments, and other forms of compensation may be provided as part of a total compensation package, in addition to a full range of medical, financial, and/or other benefits. For more information, visit: https://www.aboutamazon.com/workplace/employee-benefits. Amazon.com is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation.#0000
US, WA, Seattle
MULTIPLE POSITIONS AVAILABLE Employer: AMAZON.COM SERVICES LLC Offered Position: Research Scientist II Job Location: Seattle, Washington Job Number: AMZ10564041 Position Responsibilities: Interact with various software and business groups to develop an understanding of their business requirements and operational processes. Utilize acquired knowledge and business judgment to build decision-supporting and operational tools to improve the bottom line. Build quantitative mathematical models to represent a wide range of supply chain, transportation and logistics systems. Implement these models and tools through the use of modeling languages and engineering code in software languages such as Java, C++, C# or C. Gather required data for analysis and mathematical model building by writing ad-hoc scripts and database queries. Perform quantitative, economic, and numerical performance analyses of these systems under uncertainty using statistical and optimization tools. Create computer simulations to support operational decision-making. Identify areas with potential for improvement and work with internal teams to generate requirements to realize improvements. Design optimal or near optimal solution methodologies to be used by in-house decision support tools and software. Create software prototypes to verify and validate the devised solutions methodologies. Integrate prototypes into production systems using standard software development tools and methodologies. 40 hours / week, 8:00am-5:00pm, Salary Range: $136,000/year to $184,000/year Amazon is a total compensation company. Dependent on the position offered, equity, sign-on payments, and other forms of compensation may be provided as part of a total compensation package, in addition to a full range of medical, financial, and/or other benefits. For more information, visit: https://www.aboutamazon.com/workplace/employee-benefits. Amazon.com is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation.#0000
US, TX, Austin
MULTIPLE POSITIONS AVAILABLE Employer: AMAZON.COM SERVICES LLC Offered Position: Data Scientist III Job Location: Austin, Texas Job Number: AMZ10593703 Position Responsibilities: Own the data science elements of various products to help with data-based decision making, product performance optimization, and product performance tracking. Work directly with product managers to help drive the design of the product. Work with Technical Product Managers to help drive the build planning. Translate business problems and products into data requirements and metrics. Initiate the design, development, and implementation of scientific analysis projects or deliverables. Own the analysis, modelling, system design, and development of data science solutions for products. Write documents and make presentations that explain model/analysis results to the business. Bridge the degree of uncertainty in both problem definition and data scientific solution approaches. Build consensus on data, metrics, and analysis to drive business and system strategy. Position Requirements: Master's degree or foreign equivalent degree in Statistics, Applied Mathematics, Economics, Engineering, Computer Science or a related field and two years of experience in the job offered or a related occupation. Employer will accept a Bachelor's degree or foreign equivalent degree in Statistics, Applied Mathematics, Economics, Engineering, Computer Science, or a related field and five years of progressive post-baccalaureate experience in the job offered or a related occupation as equivalent to the Master's degree and two years of experience. Must have one year of experience in the following skills: (1) building statistical models and machine learning models using large datasets from multiple resources; (2) building complex data analyses by leveraging scripting languages including Python, Java, or related scripting language; and (3) communicating with users, technical teams, and management to collect requirements, evaluate alternatives, and develop processes and tools to support the organization. Amazon.com is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation. 40 hours / week, 8:00am-5:00pm, Salary Range $159,200/year to $215,300/year. Amazon is a total compensation company. Dependent on the position offered, equity, sign-on payments, and other forms of compensation may be provided as part of a total compensation package, in addition to a full range of medical, financial, and/or other benefits. For more information, visit: https://www.aboutamazon.com/workplace/employee-benefits. #0000
US, CA, Pasadena
The Amazon Web Services (AWS) Center for Quantum Computing (CQC) in Pasadena, CA, is hiring an Applied Scientist to lead integration research and development for superconducting quantum devices. You will design, model, and test the chip-to-chip, chip-to-package, and signal-delivery interfaces that let superconducting quantum computers scale, and you will characterize integrated assemblies at cryogenic temperatures to learn how integration choices affect qubit performance, yield, and manufacturability. You will join a multi-disciplinary team of theoretical and experimental physicists, materials scientists, and hardware and software engineers working at the forefront of quantum computing. You will carry integration technologies from concept through hardware demonstration, build the test standards and automation that make results reproducible across devices, and collaborate with fabrication, hardware, and circuit design engineers to remove the dominant sources of yield loss. You will also contribute to multi-year roadmap planning for integration and scaling capabilities and help establish the scientific foundation for this capability area within the Device team. Key job responsibilities In this role you will develop the integration technologies that let superconducting quantum computers scale. Your work will span four connected areas: - Integration research and development: Design, model, and test integration solutions for superconducting quantum devices, including chip-to-chip, chip-to-package, and signal-delivery interfaces. Build electromagnetic, thermal, and mechanical models to predict and optimize performance, and characterize integrated assemblies at cryogenic temperatures. - Scaling technologies: Develop and evaluate approaches to scaling superconducting quantum hardware against yield, reusability, manufacturability, and scalability, and carry the most promising ones from concept through hardware demonstration. - Test standards and automation: Define test standards that give rigorous, reproducible measurements of how integration affects qubit coherence, loss, and crosstalk. Automate data collection and reporting against these standards so characterization scales across devices and configurations. - Yield and manufacturability: Work with fabrication, hardware, printed circuit board, and circuit design engineers to identify and remove the largest sources of yield loss and integration uncertainty, and to design processes and components that are reused across device generations and manufactured at scale. About the team The Device team at the AWS Center for Quantum Computing designs, builds, and measures the superconducting quantum devices at the heart of our processors, and develops the integration and packaging that turn individual devices into working systems. You will work alongside physicists, materials scientists, and hardware and software engineers across design, fabrication, measurement, and theory, in a lab built for rapid iteration from concept to cryogenic test. Over the coming years the team's focus is making superconducting quantum hardware scale, and this role establishes integration as a scientific capability area within that effort. Why AWS? Amazon Web Services (AWS) is the world's most comprehensive and broadly adopted cloud platform. We pioneered cloud computing and never stopped innovating — that's why customers from the most successful startups to Global 500 companies trust our robust suite of products and services to power their businesses. AWS Utility Computing (UC) provides product innovations — from foundational services such as Amazon's Simple Storage Service (S3) and Amazon Elastic Compute Cloud (EC2), to consistently released new product innovations that continue to set AWS's services and features apart in the industry. As a member of the UC organization, you'll support the development and management of Compute, Database, Storage, Internet of Things (IoT), Platform, and Productivity Apps services in AWS. Within AWS UC, Amazon Dedicated Cloud (ADC) roles engage with AWS customers who require specialized security solutions for their cloud services. Inclusive Team Culture: Here at AWS, it's in our nature to learn and be curious. Our employee-led affinity groups foster a culture of inclusion that empower us to be proud of our differences. Ongoing events and learning experiences, including our Conversations on Race and Ethnicity (CORE) and AmazeCon (diversity) conferences, inspire us to never stop embracing our uniqueness. Diverse Experiences: AWS values diverse experiences. Even if you do not meet all of the preferred qualifications and skills listed in the job description, we encourage candidates to apply. If your career is just starting, hasn't followed a traditional path, or includes alternative experiences, don't let it stop you from applying. Mentorship & Career Growth: We're continuously raising our performance bar as we strive to become Earth's Best Employer. That's why you'll find endless knowledge-sharing, mentorship and other career-advancing resources here to help you develop into a better-rounded professional. Work/Life Balance: We value work-life harmony. Achieving success at work should never come at the expense of sacrifices at home, which is why we strive for flexibility as part of our working culture. When we feel supported in the workplace and at home, there's nothing we can't achieve in the cloud.
US, VA, Arlington
PXT Central Science is seeking an exceptional Data Scientist to join our team. The ideal candidate will thrive in a dynamic, multifaceted role where you'll translate complex business challenges into rigorous quantitative frameworks, extract actionable insights from structured and unstructured datasets, and architect science-backed, scalable solutions that elevate the experience of our 1 million+ employees worldwide. If you're energized by the opportunity to apply data science to our mission of making Amazon Earth's Best Employer, we want to hear from you. Key job responsibilities • Own the design, development, and maintenance of scalable models and prototypes leveraging statistical, machine learning, or GenAI methodologies to enhance employee experience. • Partner with scientists, engineers, and product leaders to solve for employee experience defects using scientific approaches, building new services and tools that deliverable measurable impact. • Author and maintain detailed technical documentation related to the projects you drive. • Communicate results to diverse audiences of varying technical background with effective writing, visualizations, and presentations • Stay current with emerging methods and technologies, and implement them strategically to amplify the team’s impact. About the team The Central Science Team within Amazon’s People Experience and Technology org (PXTCS) uses economics, behavioral science, statistics, machine learning, and Generative AI to proactively identify mechanisms and process improvements which simultaneously improve Amazon and the lives, well-being, and the value of work to Amazonians. We are an interdisciplinary team, which combines the talents of science, engineering, and UX to develop and deliver solutions that measurably achieve this goal.
US, NY, New York
Prime Video is a first-stop entertainment destination offering customers a vast collection of premium programming in one app available across thousands of devices. Prime members can customize their viewing experience and find their favorite movies, series, documentaries, and live sports – including Amazon MGM Studios-produced series and movies; licensed fan favorites; and programming from Prime Video subscriptions such as Apple TV+, HBO Max, Peacock, Crunchyroll and MGM+. All customers, regardless of whether they have a Prime membership or not, can rent or buy titles via the Prime Video Store, and can enjoy even more content for free with ads. Are you interested in shaping the future of entertainment? Prime Video's technology teams are creating best-in-class digital video experience. As a Prime Video team member, you’ll have end-to-end ownership of the product, user experience, design, and technology required to deliver state-of-the-art experiences for our customers. You’ll get to work on projects that are fast-paced, challenging, and varied. You’ll also be able to experiment with new possibilities, take risks, and collaborate with remarkable people. We’ll look for you to bring your diverse perspectives, ideas, and skill-sets to make Prime Video even better for our customers. With global opportunities for talented technologists, you can decide where a career Prime Video Tech takes you! Key job responsibilities The science challenges include: - Heterogeneous customer responses. The same customer responds differently to ads depending on what they’re watching, how engaged they are, and their broader streaming context. We need to understand and predict this variation to make better advertising decisions in real time. - Signal sparsity in streaming. Unlike search or retail advertising, streaming offers no clicks, no conversions, and limited direct feedback. We must develop creative approaches to infer customer preferences, intent, and tolerance from indirect behavioral signals. - Personalization at scale. A one-size-fits-all ad experience leaves value on the table for both customers and the business. We build systems that adapt ad load, placement, and content to individual viewers across 100M+ customers and 100k+ titles. - Small effects, large variance. Ad interventions typically produce 0.1-2% shifts in engagement, effects easily overwhelmed by natural behavioral variance. Measuring, attributing, and optimizing these small signals requires rigorous experimental design and causal methodologies. - Competing objectives. Revenue, customer engagement, long-term retention, and advertiser value are in tension. We develop principled frameworks to navigate these tradeoffs and optimize for sustainable outcomes rather than any single metric. A day in the life - Lead the research and development of ML models that personalize advertising decisions for 100M+ customers across 100k+ titles, with production deployment in mind - Develop deep learning architectures (multi-task learning, embedding-based representations) for customer behavior prediction at scale - Design and analyze large-scale A/B experiments, applying causal inference techniques to measure and optimize the impact of ad strategies on customer engagement and monetization - Partner with engineering to ensure models meet production latency and scalability requirements - Collaborate with product managers to frame business problems as tractable ML problems and translate findings into product decisions - Shape the team’s scientific roadmap, identifying high-impact research directions About the team Prime Video Ads is pioneering the use of Generative AI, Deep Learning and Causal ML to empower the next generation of Ads experience in Prime Video. Our vision is to build the world’s most valuable and differentiated ad-supported streaming platform: one where every ad is relevant to the viewer, and every dollar spent by an advertiser delivers measurable results. We are seeking lead applied scientists who have demonstrated experience in advancing state of the art ML algorithms, particularly in Personalization, Relevance, Casual Inference and Gen-AI areas. Your role will be to deliver these innovations as production-ready systems at Amazon scale.
US, WA, Redmond
Amazon Leo is building a low Earth orbit satellite broadband network to deliver fast, affordable connectivity to customers around the world, including those underserved by traditional infrastructure. Operating a live constellation and its ground network at scale means service disruptions are an inevitable reality, and how quickly we detect them, understand them, and prevent them from recurring defines the trust we earn as a connectivity provider. We are seeking a talented Senior Data Scientist to build the next generation of observability and availability science for the Leo network. Leo already detects outages, notifies affected customers, and restores service. The frontier is prediction and prevention: reasoning over complex network and ground-segment telemetry to autonomously detect anomalies before they reach customers, attribute root cause automatically across a sprawling event surface, and rationalize a large, still-maturing metric set so responders trust what the system tells them. In this role you will work backwards from the needs of customers and operations teams. You will own the scientific direction end to end, partnering closely with network engineers, the Network Operations Center, and the Incident Management team, and the Software and Networking orginization to translate a long-term science vision into concrete research and delivery roadmaps. Export Control Requirement: Due to applicable export control laws and regulations, candidates must be a U.S. citizen or national, U.S. permanent resident (i.e., current Green Card holder), or lawfully admitted into the U.S. as a refugee or granted asylum. Key job responsibilities * Own the observability and availability science end to end, from long-term vision to research roadmap to production delivery, working backwards from the needs of customers, the NOC, and network engineering. * Develop time-series and graph-based models with dynamic baselining that detect network and ground-segment anomalies before they cross the High Severity threshold, enabling proactive investigation rather than reactive response. * Reason over complex, multi-system telemetry to attribute cause automatically during live HSEs and LSEs, so Call Leaders and engineers spend time mitigating, not hunting. * Provide the science behind the network's bucketed availability decomposition (geometric and satellite coverage, ground and network segments, customer terminals, payload, and planning), attributing every movement to a true cause rather than noise, so leadership steers on signal and engineering effort lands where it actually moves availability and throughput. * Deliver the models, analyses, and metrics that feed the daily system readiness and validation review consumed by senior leadership and product and engineering, turning availability and throughput data into a defensible, evidence-based read on progress against GA entry criteria. * Own the recurring, fleet-wide "hot topic" investigations that span ASIC, firmware, payload, network, and planning, separating correlated symptoms from true root cause across thousands of nodes, quantifying how much availability or throughput each is costing, and answering whether it is a launch-blocking risk and whether it is improving. * Deliver alert correlation and deduplication, false-positive reduction, and metric-quality scoring so responders trust what the system tells them, directly supporting a large, still-maturing KPI set. * Tie your science to the unit the org actually measures: reduce detection and attribution time, raise availability against the network's bar, and increase the share of events prevented before customer impact. * Partner day to day with the Network Operations Center, Network Engineering, and the Incident Management (Call Leader) program; sit on live HSE/LSE bridges as the science partner and turn event learnings into model improvements. * Feed COEs and RCAs back into detection and attribution so recurring signatures are caught earlier, and help design and score gamedays that stress-test both the network and your models against real, closed events. * Build and deploy production pipelines on AWS (e.g., SageMaker, Bedrock, Glue) with model validation, monitoring, drift detection, and explainability, because operations teams must be able to trust and act on model output under pressure. * Set data-science standards, mentor scientists, and drive best practices across the Leo networking organization as the science function scales. * Produce explainable output and written narratives that let responders and leaders three or more levels up act on your models during and between events. A day in the life You might start by reviewing overnight anomaly detections and availability movements, join an HSE bridge as the science partner by midday, and spend the afternoon refining root-cause attribution models or presenting availability findings at the readiness review. About the team Amazon Leo Networking Software builds and operates the software behind Leo's satellite and ground network. Our science function partners with the NOC, Network Engineering and Incident Management to make the network more available and resilient, moving from reacting to outages to predicting and preventing them. As a Data Scientist III, you'll help shape that function as it grows.
US, CA, Santa Clara
Amazon Customer Service (CS) org’s mission is to make customer service interactions so good they become the primary reason customers choose Amazon. Customer Experience Products (CXP) supports this vision by building customer-facing solutions customers prefer because they work - reducing effort, shortening time to resolution, and building trust with every interaction. Our science team evaluates AI and Large Language Models (LLM)-driven solutions for both consumer-facing and CS associates-facing AI applications. We evaluate multi-modal, multi-turn, goal-oriented dialog systems (e.g. chatbot, voicebot) that can resolve customer issues and provide informational answers grounded by Amazon policy and help articles, at Amazon scale across multiple languages, intents and locales. These fine-tuned LLM solutions are designed to adapt to changing company policies and automate solutions to customer problems with conversational intelligence. Additionally, we're exploring with the partner associate-facing org on defining quality of LLM capabilities to improve the productivity of associates. Key job responsibilities - Lead development of Customer Service specific multi modal foundational models that improve chat and voice experiences. - Establish and evaluate rigorous testing and validation protocols for AI-driven systems. - Lead ongoing quality improvements for launched AI solutions - Collaborate closely with Product Managers, UX Designers, Applied Scientists, and Software Development Engineers. - Practice thought leadership and contribute to the overall science vision and strategic direction for AI-driven customer experience transformation initiatives. - Stay at the forefront of AI and machine learning technologies.
CA, BC, Vancouver
Building AI first solutions in a complex regulated space at Amazon scale. Would you like to work on one of the world's largest product classification and Tax computation engine? Or perhaps teaching AI to understanding nuances of domain specific language is what piques your interest? We are responsible for the tax calculation platform, providing the core services that calculate indirect taxes (sales tax and VAT) for all Amazon sales, physical and digital, globally. We are responsible for classifying billions of products in Amazon’s catalogue, calculating accurate Tax amount for every transaction that happens on Amazon’s global platform. Our engineering services are relied upon globally, built to be resilient across some of highest peak traffic demands in the world, while constantly evolving with the state of the art technology. We use language models at scale for Tax classification of the diverse products in Amazon catalogue and comprehending nuances of legislations and legal parlour.Our challenges include staying on top of the complex and fast-changing global tax legislations while ensuring computing calculations correctly and quickly, thousands of times a second, and each one needs to be accurate. Applied Scientists in Tax Engine works closely with Tax specialists, Product team , SDEs and Business teams - tackling a growing portfolio of core-science problems that includes balance of predictive and generative AI -including language comprehension, causal reasoning and active learning. As an Applied scientist, you will provide machine learning expertise to the team that helps increase the accuracy of Tax classification based on product information in Amazon catalogue. You will build various machine learning models to predict accuracy of tax classification, work with large reasoning models to reason with large volumes of system data and expert artefacts to teach domain intuition to AI. You will on variety of legislative changes acting as causal determinants to changes in our systems. You will help us innovate different ways to enhance for our global customers of Amazon.You will need to be entrepreneurial, work in a highly collaborative environment with SDEs, Product managers and businesses. We like to move fast, experiment, iterate and then scale quickly, thoughtfully balancing speed and quality. Whether you're passionate about taking your scientific solves to global reach and scale or one who seeks to solve complex problems under high scrutiny of an audited field, Tax Engine’s scope could offer you an intersting challenge to solve.
US, NY, New York
MULTIPLE POSITIONS AVAILABLE Employer: AMAZON.COM SERVICES LLC Offered Position: Data Scientist III Job Location: New York, New York Job Number: AMZ10444216 Position Responsibilities: Own the data science elements of various products to help with data-based decision making, product performance optimization, and product performance tracking. Work directly with product managers to help drive the design of the product. Work with Technical Product Managers to help drive the build planning. Translate business problems and products into data requirements and metrics. Initiate the design, development, and implementation of scientific analysis projects or deliverables. Own the analysis, modelling, system design, and development of data science solutions for products. Write documents and make presentations that explain model/analysis results to the business. Bridge the degree of uncertainty in both problem definition and data scientific solution approaches. Build consensus on data, metrics, and analysis to drive business and system strategy. Position Requirements: Master's degree or foreign equivalent degree in Statistics, Applied Mathematics, Economics, Engineering, Computer Science or a related field and two years of experience in the job offered or a related occupation. Employer will accept a Bachelor's degree or foreign equivalent degree in Statistics, Applied Mathematics, Economics, Engineering, Computer Science, or a related field and five years of progressive post-baccalaureate experience in the job offered or a related occupation as equivalent to the Master's degree and two years of experience. Must have one year of experience in the following skills: (1) building statistical models and machine learning models using large datasets from multiple resources; (2) building complex data analyses by leveraging scripting languages including Python, Java, or related scripting language; and (3) communicating with users, technical teams, and management to collect requirements, evaluate alternatives, and develop processes and tools to support the organization. Amazon.com is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation. 40 hours / week, 8:00am-5:00pm, Salary Range $175,100/year to $236,900/year. Amazon is a total compensation company. Dependent on the position offered, equity, sign-on payments, and other forms of compensation may be provided as part of a total compensation package, in addition to a full range of medical, financial, and/or other benefits. For more information, visit: https://www.aboutamazon.com/workplace/employee-benefits.#0000