Automated reasoning at Amazon: A conversation

To mark the occasion of the eighth Federated Logic Conference (FloC), Amazon’s Byron Cook, Daniel Kröning, and Marijn Heule discussed automated reasoning’s prospects.

The Federated Logic Conference (FLoC) is a superconference that, like the Olympics, happens every four years. FLoC draws together 12 distinct conferences on logic-related topics, most of which meet annually. The individual conferences have their own invited speakers, but FLoC as a whole has several plenary speakers as well.

At the last FLoC, in 2018, one of those plenary speakers was Byron Cook, who leads Amazon’s automated-reasoning group, and he was introduced by Daniel Kröning, then a professor of computer science at the University of Oxford

Byron Cook's keynote at FLoC 2018
With introduction by Daniel Kröning.

“What makes me so proud that Byron is here,” Kröning said, is “he’s now at Amazon, and he’s going to run the next Bell Labs, he’s going to run the next Microsoft Research, from within Amazon. My prediction is that — not 10 years but 16 years; remember, it’s multiples of four — 16 years from now you’ll be at a FLoC, and you’ll hear these stories about the great thing that Byron Cook built up at Amazon Web Services. And we’ll speak about it in the same tone as we’re now talking about Bell Labs and Microsoft Research.”

In the audience at the talk was Marijn Heule, a highly cited automated-reasoning researcher who was then at the University of Texas.

“I hadn't met Marijn, though I had heard about him from a couple other people and thought I should talk to him,” Cook says. “And then Marijn found me at the banquet after the talk and was like, ‘I want a job.’”

AR scientists.png
L to R: Amazon vice president and distinguished scientist Byron Cook; Amazon Scholar Marijn Heule; Amazon senior principal scientist Daniel Kröning.

Heule is now an Amazon Scholar who divides his time between Amazon and his new appointment at Carnegie Mellon University. Kröning, too, has joined Amazon as a senior principal scientist, working closely with Cook’s group.

As 2022’s FLoC approached, Cook, Kröning, and Heule took some time to talk with Amazon Science about the current state of automated-reasoning research and its implications for Amazon customers.

Related content
Meet Amazon Science’s newest research area.

Amazon Science: The conference name has the word “logic” in it. Does FLoC deal with other aspects of logic, or is logic coextensive with automated reasoning now?

Byron Cook: It’s about the intersection of logic and computer science. Automated reasoning is one dimension of that intersection.

Daniel Kröning: Traditionally, FLoC is split into two halves, with the first half more theoretical and the second half more applied.

Cook: One of the things about automated reasoning is you're on the bleeding edge of what is even computable. We're often working on intractable or undecidable problems. So people automating reasoning are really paying attention to both the applied and the theoretical.

AS: I know Marijn is concentrating on SAT solvers, and SAT is an intractable problem, right? It’s NP-complete?

Marijn Heule: Yes, but you can also use these techniques to solve problems that go beyond NP. For example, solvers for SAT modulo theories, called SMT. I even have a project with one student trying to solve the famous Collatz conjecture with these tools.

Collatz-27.png
The Collatz conjecture posits that any integer will be transformed into the integer 1 through iterative application of two operations: n/2 and 3n+1. This figure shows a "Collatz cascade" of possible transitions from 27 to 1 using a set of seven symbols, which can be interpreted as simple calculations, and 11 rules for transforming those symbols into symbols consistent with the Collatz operations. At top right are the symbol rewrite rules; at bottom left is a blowup of part of the cascade, illustrating sequences of rewrites that yield the number 425 and its transformation through Collatz operations.

Kröning: SAT is now the inexpensive, easy-to-solve workhorse for really hard problems. People still have it in their heads that SAT equals NP hard, therefore difficult to solve or impossible to solve. But for us, it's the lowest entry point. On top of SAT, we build algorithms for solving problems that are way harder.

Cook: One of the tricks of the trade is abstraction, where you take a problem that's much, much bigger but represent it with something smaller, where classes of questions you might ask about the smaller problem imply that the answer also holds for the bigger problem. We also have techniques for refining the abstractions on demand when the abstraction is losing too much information to answer the question. Often we can represent these abstractions in tools for SAT.

Related content
Distributing proof search, reasoning about distributed systems, and automating regulatory compliance are just three fruitful research areas.

Marijn’s work on the Collatz conjecture is a great example of this. He has done this amazing reduction of Collatz to a series of SAT questions, and he's tantalizingly close to solving it because he's got one decidable problem to go — and he's the world expert on solving those problems. [Laughs]

Heule: Tantalizingly close but also so far away, right? Because this problem might not be solvable even with a million cores.

Cook: But it's still decidable. And one of the thresholds is that NP, PSpace, all these things, they're actually decidable. There are questions that are undecidable — and we work on those, too. When a problem is undecidable, it means that your tool will sometimes fail to find an answer, and that's just fundamental: there are no extra computers you could use ever to solve that. The halting problem is a great example of that.

Heule: For these kinds of problems, you're asking the question “Is there a termination argument of this kind of shape?” And if there is one, you have your termination argument. If there is no termination argument of that shape, there could be one of another shape. So if the answer is SAT [satisfiable], then you're happy because you’ve solved the problem. If the answer is no, you try something else.

Cook: It's really, really exciting. In Amazon, we're building these increasingly powerful SAT solvers, using the power of the cloud and distributed systems. So there's no better place for Marijn to be than at Amazon.

Related content
ICSE paper presents techniques piloted by Amazon Web Services’ Automated Reasoning team.

AS: Daniel, could we talk a little bit about your research?

Kröning: What I'm looking at right now is reasoning about the cloud infrastructure that performs remote management of EC2 instances — how to secure that in a way that is provable. You also want to do that in a way that is economical.

Cook: One of the things that Daniel's focusing on is agents. We have pieces of software that run on other machines, like EC2 instances, agents for telemetry or for control, and you give them power to take action on your behalf on your machine. But you want to make sure that an adversary doesn't trick those agents into doing bad things.

Correct software

AS: I know that, commercially, formal methods have been used in hardware design and transportation systems for some time. But it seems that they’re really starting to make inroads in software development, too.

The storage team is able to write code that otherwise they might not want to deploy because they wouldn't be as confident about it, and they're deploying four times as fast. It was an investment in agility that's really paid off.
Byron Cook

Cook: The thing we've seen is it's really by need. The storage team, for example, is able to be much more agile and be much more aggressive in the programs that they write because of the formal methods. They're able to write code that otherwise they might not want to deploy because they wouldn't be as confident about it, and they're deploying four times as fast. It was an investment in agility that's really paid off.

Kröning: There are actually a good number of stories wherein engineering teams didn't dare to roll out a particular feature or design revision or design variant that offers clear benefits — like being faster, using less power — because they just couldn't gain the confidence that it's actually right under all circumstances.

Heule: The interesting thing is that you even see this now in tools. Now we have produced proofs from the tools, and people start implementing features that they wouldn't dare have in the past because they were not clear that they were correct. So the solvers get faster and more complex because we now can check the results from the tools and to have confidence in their correctness.

Related content
SOSP paper describes lightweight formal methods for validating new S3 data storage service.

Cook: Yeah, I wanted to double down on that point. There’s a distinction in automated reasoning between finding a proof and checking your proof, and the checking is actually relatively easy. It's an accounting thing. Whereas finding the proof is an incredibly creative activity, and the algorithms that find proofs are mind-blowing. But how do you know that the tool that found the proof is correct? Well, you produce an auditable artifact that you can check with the easy tool.

SAT in the cloud

AS: What are you all most excited about at this year’s FLoC?

Cook: The SAT conference is at FLoC, and there will be the SAT competition results, and one of the things I'm really excited about is the cloud track. Automated reasoning has really moved into the cloud, and the past couple years running the cloud track has really blown the doors off what's possible. I'm expecting that that will be true again this year.

SAT results.png
The results of the top-performing cloud-based, parallel, and sequential SAT solvers in this year's SAT competition, whose results were presented at FLoC. The curves show the number of problems (y-axis) in the SAT competition's anniversary problem set — which aggregates all 5,355 problems presented in the competition's 20-year history — that a given solver could solve in the allotted time (x-axis).

Heule: This is the first year that Amazon is running both the parallel track and the cloud track, and the cloud track was only possible because of Amazon. Before that, there was no way we had the resources to run a cloud track. In the cloud track, every solver-benchmark combination is run on 1,600 cores. And this year is extra special because it's 20 years of SAT, and we have a single anniversary track and all the competitions that were run in the past are in there. That is 5,355 problems, and all the solvers are running on this.

Cook: Wow.

Heule: I'm also excited to see the results. We have seen in the last year and the year before that the cloud solver can, say, solve in 100 seconds as much as the sequential solvers can do in 5,000 seconds. The user doesn't have to wait for four hours but just for four minutes

Cook: And that raises all boats because, as we mentioned earlier, everything is reduced to SAT. If the SAT solvers go from one hour to one minute, that's really game changing. That means a whole other set of things you can do.

What has been clear for a while but continues to be true is there's some sort of Moore's-law thing happening with SAT. You fix the same hardware, the same benchmarks, and then run all the tools from the past 20 years, and you see every year they're getting dramatically better. What's also really amazing is that in many ways the tools are getting simpler.

LH: Are the simplicity and efficiency two sides of the same coin? Understanding the problems better helps you find a simpler solution, which is more efficient?

Cook: Yeah, but it’s also the point that Marijn made that because the tools produce auditable proofs that you can check independently, you can do aggressive things that we were scared to do before. Often, aggressive is much simpler.

Related content
Automated-reasoning method enables the calculation of tight bounds on the use of resources — such as computation or memory — that results from code changes.

Heule: It's also the case that we now understand there are different kinds of problems, and they need different kinds of heuristics. Solvers are combining different heuristics and have phases: “Let's first try this. Let's also try that.” And the code involved in changing the heuristics is very small. It's just changing a couple of parameters. But if you notice, okay, this set of heuristics works well for this problem, then you kind of focus more on that.

Cook: One of the things a SAT solver does is make decisions fast. It just makes a bunch of choices, and those choices won't work out, and then it spends some time to learn lessons why. And then it has a very efficient internal database for managing what has been learned, what not to do in the future. And that prunes the search space a lot.

One of the really exciting things that's happening in the cloud is that you have, say, 1,000 SAT solvers all running on the same problem, and they're learning different things and can share that information amongst them. So by adding 5,000 more solvers, if you can make the communication and the lookup efficient between them, you're really off to the races.

The other thing that's quite neat about that is the point that Marijn is making: it's becoming increasingly clear that there are these fundamental building blocks, and for different kinds of problems, you would want to use one kind of Lego brick versus a different kind of Lego brick. And the cloud allows you to run them all but then to share the information between them.

Iterated SAT solver.png
In "Migrating solver state", Heule and his colleagues show that passing modified versions of a problem between different solvers can accelerate convergence on a solution.

Heule: We have an Amazon paper at FLoC with some very cool ideas. If you run things in the cloud, you sometimes have a limited time window where you have to solve them, and otherwise it stops. You started with a certain problem, the solver did some modifications, and now we have a different problem. Initially we just tested, Okay, can we stop the solver and then store the modified problem somewhere and continue later, in case we need more time than we allocated initially? And then we can continue solving it.

But the interesting thing is that if you give the modified problem to another solver, and you give it, say, a couple of minutes, and then it stores the modified problem, and you give it to another solver, it actually really speeds things up. It turns out to solve the most instances from everything that we tried.

AS: Do you do that in a principled way, or do you just pick a new solver randomly?

Related content
In a pilot study, an automated code checker found about 100 possible errors, 80% of which turned out to require correction.

Heule: The thing that turned out to work really well is to take two top-tier solvers and just Ping-Pong the problem among them. This functionality of storing and continuing search requires some work, so that implementing it in, say, a dozen solvers would require quite some work. But it would be a very interesting experiment.

AS: I’m sure our readers would love to know the result of that experiment!

Well, thank you all very much for your time. Does anyone have any last thoughts?

Cook: I think I speak for the thousands of others who are attending FLoC: we are ready to having our minds blown, just as we did in 2018. Many of the tools and theories presented by our scientific colleagues at this year’s FLoC will challenge our current assumptions or spark that next big insight in our brains. We will also get to catch up with old friends that we’ve known for around 20 years and meet new ones. I’m particularly excited to meet the new generation of scientists who have entered the field, to see the world afresh through their eyes. This is such an amazing time to be in the field of automated reasoning.

Research areas

Related content

US, WA, Seattle
Are you excited about using economic reasoning to shape decisions that affect millions of customers? Do you want to tackle ambiguous, high-impact problems where your analysis directly influences strategy and investment? We are looking for a Senior Economist to join Amazon and drive economic solutions for complex business challenges. In this role, you will define and prioritize analytical approaches, build rigorous econometric models, and translate findings into actionable recommendations for leadership. You will partner with engineers, product managers, scientists, and business leaders to deliver measurable outcomes. If you thrive on turning unstructured problems into clear, data-driven strategies, this is the role for you. Key job responsibilities - Lead the design, implementation, and delivery of economic analyses for complex or ambiguous business problems, providing strategic direction that informs organizational goals and investment decisions. - Develop and apply econometric models—including causal inference, structural estimation, and forecasting methods—choosing the right approach for each problem and validating results through rigorous testing. - Collaborate with cross-functional partners including engineers, product managers, and applied scientists to implement scalable solutions and translate economic insights into product and business improvements. - Communicate findings through clear narratives and technical documents for audiences ranging from peer economists to senior leaders, building consensus and influencing key decisions. - Evaluate and integrate generative AI tools where appropriate to augment research productivity, while mentoring other economists and raising the bar on analytical best practices across the team. A day in the life You might start your morning reviewing model performance metrics and refining an econometric specification before a working session with engineers on data pipeline requirements. After lunch, you could present preliminary findings from a causal analysis to business stakeholders, fielding questions and aligning on next steps. Later in the afternoon, you join a peer review for a fellow economist's research design, offering constructive feedback. Throughout the day, you balance hands-on analytical work with strategic conversations that connect your research to real business outcomes. About the team Our team of economists partners across Amazon to solve some of the company's most consequential business questions. We combine deep expertise in economic theory and applied econometrics with access to rich, large-scale data to deliver insights that shape long-term strategy. We value intellectual curiosity, collaborative problem-solving, and a commitment to rigorous science. If you are looking for an environment where your economic thinking directly drives impact at scale, we would love to hear from you.
US, WA, Redmond
Amazon Leo is building a low Earth orbit satellite broadband network to bring fast, affordable connectivity to customers worldwide. We're looking for a Senior Applied Scientist to invent the machine learning that will let the network predict, explain and prevent its own failures. You'll work on open research problems at large scale. These include anomaly detection on non-stationary, high-dimensional telemetry, causal root-cause inference over constantly changing satellite and ground network graphs, and trustworthy ML for operators making decisions under pressure. You'll own the science agenda, from forming hypotheses and designing new models to putting them into production systems that protect customer connectivity. 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 Define and lead a multi-year applied research agenda in network observability, anomaly prediction and availability modeling. Invent new time-series and spatio-temporal graph models, such as graph neural networks (GNNs), foundation models and dynamic baselining, to predict network and ground-segment failures before customers are affected. Develop causal inference and causal discovery methods for automated root-cause attribution across a large, interdependent event surface. Research probabilistic models that break availability and throughput into causal drivers (coverage, ground, network, terminal, payload, planning) and separate true signal from noise. Advance methods for alert correlation, deduplication, uncertainty quantification and metric-quality scoring. Explore large language models (LLMs) and agentic approaches for reasoning over telemetry, logs and incident narratives. Design rigorous offline and online evaluation, including backtesting against real closed incidents and gameday-based benchmarks. Take research into production on AWS (SageMaker, Bedrock), working with engineers on scalability, drift detection and explainability. Publish at top-tier venues, file patents and raise the scientific bar through mentoring and design reviews. A day in the life You might start the morning reading new literature on causal discovery and sketching how it applies to Leo's network graph. Later you could review a scientist's experiment design and debate the best evaluation methodology. During an event, you would test your attribution model live on the incident call. You might finish the day prototyping a new spatio-temporal anomaly detector and backtesting it against last quarter's incidents. About the team Amazon Leo Networking Software builds the software behind Leo's satellite and ground network. Our applied science team works on hard, open-ended problems with real-world impact. We value scientific rigor, curiosity and ownership, and we encourage publishing and contributing to the research community. As a Senior Applied Scientist, you'll help build this science function from the ground up and set the technical bar for the scientists who join after you.
US, WA, Bellevue
We are seeking a Senior Applied Scientist to push the state of the art in robot learning and help bring capable robots out of the lab and into everyday homes at Fauna Robotics. In this role, you will design novel algorithms across perception, manipulation, sim2real transfer, reinforcement learning, and multi-task learning, closing the gap between research and real-world deployment. You will work where robotics research meets production systems. You'll combine hands-on technical work with scientific leadership to guide the team toward the most promising directions. Working closely with software engineers and scientists, you will turn new ideas into robust solutions for dynamic, unstructured home environments. Homes bring challenges that structured settings like warehouses don't, and consumer-grade sensors and actuators have real limits, so this work calls for creative, out-of-the-box approaches. Key job responsibilities - Lead technical initiatives in areas such as robotics foundation models, reinforcement learning and manipulation - - Design experiments to identify the limitations of current state-of-the-art models and developing new models or techniques that can surpass these models - Design and implement novel deep learning architectures that push the boundaries of what robots can understand and accomplish - Mentor fellow scientists while maintaining strong individual technical contributions - Collaborate with engineering teams to optimize and scale models for real-world applications - Influence technical decisions and implementation strategies within your area of focus - Experienced in communicating complex technical work to a non-technical audience - The ability to work with minimal guidance, be proactive and to handle ambiguity and the challenge of quickly evolving goals About the team Fauna Robotics, an Amazon company, is building capable, safe, and genuinely delightful robots for everyday life. Our goal is simple: make robots people actually want to live and interact with in everyday human spaces. We believe that future won’t arrive until building for robotics becomes far more accessible. Today, too much effort is spent reinventing the fundamentals. We’re changing that by developing tightly integrated hardware and software systems that make it faster, safer, and more intuitive to create real-world robotic products. Our work spans the full stack: mechanical design, control systems, dynamic modeling, and intelligent software. The focus is not just functionality, but experience. We’re building robots that feel responsive, expressive, and genuinely useful. At Fauna, you’ll work at the frontier of this space, helping define how robots move, manipulate, and interact with people in natural environments. It’s an opportunity to solve hard problems across hardware and software with a team focused on making robotics accessible and joyful to build. If you care about making robotics real for everyone and building systems that are as delightful as they are capable, we’re interested in hearing from you.
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, 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, 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, WA, Bellevue
Within Amazon's North American Operations, we are tackling mission-critical problems to optimize network operations. We build complex causal, forecasting, and optimization models to better understand, predict, and govern processes across hundreds of warehouses. We are seeking a Data Scientist to drive the development of these models, deeply integrate with business stakeholders, and leverage generative AI to more intelligently build for our customers. Check out the hiring managers background - https://www.linkedin.com/pulse/from-nuclear-power-network-intelligence-david-plews-brittany-sdkbe/ Key job responsibilities - Interfacing with stakeholders to understand business problems and translating these into high and low-level design documents. - Building machine learning, optimization, and causal inference models that target business-specific challenges. - Sourcing, cleaning, and analyzing large and complex data sets associated with warehouse operations to provide business-critical insights at scale. - Leveraging AI-based solutions to drive development and provide nuanced insights into model outcomes. A day in the life You will be a single-threaded owner of projects within Amazon Operations, owning project scoping, stakeholder engagement, data enablement, and model development. Your work will be directly consumed by partner teams throughout Operations and Finance, and hence a strong sense of ownership is required to educate both technical and non-technical stakeholders on developed science solutions. About the team Our team is comprised of Data Scientists, Applied Scientists, and Economists, and are trusted to make data-driven solutions that power operations decision making. We strive to build consensus on divisive topics through clear presentation of problem statements paired with deep analyses that gain visibility throughout all levels of the organization from operators to executives.
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.
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, 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