Physics-constrained machine learning for scientific computing

Amazon researchers draw inspiration from finite-volume methods and adapt neural operators to enforce conservation laws and boundary conditions in deep-learning models of physical systems.

Commercial applications of deep learning have been making headlines for years — never more so than this spring. More surprisingly, deep-learning methods have also shown promise for scientific computing, where they can be used to predict solutions to partial differential equations (PDEs). These equations are often prohibitively expensive to solve numerically; using data-driven methods has the potential to transform both scientific and engineering applications of scientific computing, including aerodynamics, ocean and climate, and reservoir modeling.

A fundamental challenge is that the predictions of deep-learning models trained on physical data typically ignore fundamental physical principles. Such models might, for instance, violate system conservation laws: the solution to a heat transfer problem may fail to conserve energy, or the solution to a fluid flow problem may fail to conserve mass. Similarly, a model’s solution may violate boundary conditions — say, allowing heat flow through an insulator at the boundary of a physical system. This can happen even when the model’s training data includes no such violations: at inference time, the model may simply extrapolate from patterns in the training data in an illicit way.

In a pair of recent papers accepted at the International Conference on Machine Learning (ICML) and the International Conference on Learning Representations (ICLR), we investigate the problems of adding known physics constraints to the predictive outputs of machine learning (ML) models when computing the solutions to PDEs.

Related content
Danielle Maddix Robinson's mathematics background helps inform robust models that can predict everything from retail demand to epidemiology.

The ICML paper, “Learning physical models that can respect conservation laws”, which we will present in July, focuses on satisfying conservation laws with black-box models. We show that, for certain types of challenging PDE problems with propagating discontinuities, known as shocks, our approach to constraining model outputs works better than its predecessors: it more sharply and accurately captures the physical solution and its uncertainty and yields better performance on downstream tasks.

In this paper, we collaborated with Derek Hansen, a PhD student in the Department of Statistics at the University of Michigan, who was an intern at AWS AI Labs at the time, and Michael Mahoney, an Amazon Scholar in Amazon’s Supply Chain Optimization Technologies organization and a professor of statistics at the University of California, Berkeley.

In a complementary paper we presented at this year’s ICLR, “Guiding continuous operator learning through physics-based boundary constraints”, we, together with Nadim Saad, an AWS AI Labs intern at the time and a PhD student at the Institute for Computational and Mathematical Engineering (ICME) at Stanford University, focus on enforcing physics through boundary conditions. The modeling approach we describe in this paper is a so-called constrained neural operator, and it exhibits up to a 20-fold performance improvement over previous operator models.

So that scientists working with models of physical systems can benefit from our work, we’ve released the code for the models described in both papers (conservation laws | boundary constraints) on GitHub. We also presented on both works in March 2023 at AAAI's symposium on Computational Approaches to Scientific Discovery.

Danielle Maddix Robinson on physics-constrained machine learning for scientific computing
A talk presented in April 2023 at the Machine Learning and Dynamical Systems Seminar at the Alan Turing Institute.

Conservation laws

Recent work in scientific machine learning (SciML) has focused on incorporating physical constraints into the learning process as part of the loss function. In other words, the physical information is treated as a soft constraint or regularization.

Related content
Hybrid model that combines machine learning with differential equations outperforms models that use either strategy by itself.

A main issue with these approaches is that they do not guarantee that the physical property of conservation is satisfied. To address this issue, in “Learning physical models that can respect conservation laws”, we propose ProbConserv, a framework for incorporating constraints into a generic SciML architecture. Instead of expressing conservation laws in the differential forms of PDEs, which are commonly used in SciML as extra terms in the loss function, ProbConserv converts them into their integral form. This allows us to use ideas from finite-volume methods to enforce conservation.

In finite-volume methods, a spatial domain — say, the region through which heat is propagating — is discretized into a finite set of smaller volumes called control volumes. The method maintains the balance of mass, energy, and momentum throughout this domain by applying the integral form of the conservation law locally across each control volume. Local conservation requires that the out-flux from one volume equals the in-flux to an adjacent volume. By enforcing the conservation law across each control volume, the finite-volume method guarantees global conservation across the whole domain, where the rate of change of the system’s total mass is given by the change in fluxes along the domain boundaries.

Flux Volume Edit-01_230525135151.jpg
The integral form of a conservation law states that the rate of change of the total mass of the system over a domain (Ω) is equal to the difference between the in-flux and out-flux along the domain boundaries (∂Ω).

More specifically, the first step in the ProbConserv method is to use a probabilistic machine learning model — such as a Gaussian process, attentive neural process (ANP), or ensembles of neural-network models — to estimate the mean and variance of the outputs of the physical model. We then use the integral form of the conservation law to perform a Bayesian update to the mean and covariance of the distribution of the solution profile such that it satisfies the conservation constraint exactly in the limit.

Related content
Learning the complete quantile function, which maps probabilities to variable values, rather than building separate models for each quantile level, enables better optimization of resource trade-offs.

In the paper, we provide a detailed analysis of ProbConserv’s application to the generalized porous-medium equation (GPME), a widely used parameterized family of PDEs. The GPME has been used in applications ranging from underground flow transport to nonlinear heat transfer to water desalination and beyond. By varying the PDE parameters, we can describe PDE problems with different levels of complexity, ranging from “easy” problems, such as parabolic PDEs that model smooth diffusion processes, to “hard” nonlinear hyperbolic-like PDEs with shocks, such as the Stefan problem, which has been used to model two-phase flow between water and ice, crystal growth, and more complex porous media such as foams.

For easy GPME variants, ProbConserv compares well to state-of-the-art competitors, and for harder GPME variants, it outperforms other ML-based approaches that do not guarantee volume conservation. ProbConserv seamlessly enforces physical conservation constraints, maintains probabilistic uncertainty quantification (UQ), and deals well with the problem of estimating shock propagation, which is difficult given ML models’ bias toward smooth and continuous behavior. It also effectively handles heteroskedasticity, or fluctuation in variables’ standard deviations. In all cases, it achieves superior predictive performance on downstream tasks, such as predicting shock location, which is a challenging problem even for advanced numerical solvers.

Examples

Conservation of mass.png
Conservation of mass can be violated by a black-box deep-learning model (here, the ANP), even when the PDE is applied as a soft constraint (here, SoftC-ANP) on the loss function, à la physics-informed neural networks (PINNs). This figure shows the variation of total mass over time for the smooth constant coefficient diffusion equation (an “easy” GPME example). The true mass remains zero, since there is zero net flux from the domain boundaries, and thus mass cannot be created or destroyed in the domain interior.
Uncertainty quantification.png
Density solution profiles with uncertainty quantification. In the “hard” version of the GPME problem, also known as the Stefan problem, the solution profile may contain a moving sharp interface in space, known as a shock. The shock here separates the region with fluid from the degenerate one with zero fluid density. The uncertainty is largest in the shock region and becomes smaller in the areas away from it. The main idea behind ProbConserv’s UQ method is to use the uncertainty in the unconstrained black box to modify the mean and covariance at the locations where the variance is largest, to satisfy the conservation constraint. The constant-variance assumption in the HardC-ANP baseline does not result in improvement on this hard task, while ProbConserv results in a better estimate of the solution at the shock and a threefold improvement in the mean squared error (MSE).
Shock position.png
Downstream task. Histogram of the posterior of the shock position computed by ProbConserv and the other baselines. While the baseline models skew the distribution of the shock position, ProbConserv computes a distribution that is well-centered around the true shock position. This illustrates that enforcing physical constraints such as conservation is necessary in order to provide reliable and accurate estimations of the shock position.

Boundary conditions

Boundary conditions (BCs) are physics-enforced constraints that solutions of PDEs must satisfy at specific spatial locations. These constraints carry important physical meaning and guarantee the existence and the uniqueness of PDE solutions. Current deep-learning-based approaches that aim to solve PDEs rely heavily on training data to help models learn BCs implicitly. There is no guarantee, though, that these models will satisfy the BCs during evaluation. In our ICLR 2023 paper, “Guiding continuous operator learning through physics-based boundary constraints”, we propose an efficient, hard-constrained, neural-operator-based approach to enforcing BCs.

Related content
Amazon quantum computing scientist recognized for ‘outstanding contributions to physics’.

Where most SciML methods (for example, PINNs) parameterize the solution to PDEs with a neural network, neural operators aim to learn the mapping from PDE coefficients or initial conditions to solutions. At the core of every neural operator is a kernel function, formulated as an integral operator, that describes the evolution of a physical system over time. For our study, we chose the Fourier neural operator (FNO) as an example of a kernel-based neural operator.

We propose a model we call the boundary-enforcing operator network (BOON). Given a neural operator representing a PDE solution, a training dataset, and prescribed BCs, BOON applies structural corrections to the neural operator to ensure that the predicted solution satisfies the system BCs.

BOON architecture full.png
BOON architectures. Kernel correction architectures for commonly used Dirichlet, Neumann, and periodic boundary conditions that carry different physical meanings.

We provide our refinement procedure and demonstrate that BOON’s solutions satisfy physics-based BCs, such as Dirichlet, Neumann, and periodic. We also report extensive numerical experiments on a wide range of problems including the heat and wave equations and Burgers's equation, along with the challenging 2-D incompressible Navier-Stokes equations, which are used in climate and ocean modeling. We show that enforcing these physical constraints results in zero boundary error and improves the accuracy of solutions on the interior of the domain. BOON’s correction method exhibits a 2-fold to 20-fold improvement over a given neural-operator model in relative L2 error.

Examples

Insulator at boundary.png
Nonzero flux at an insulator on the boundary. The solution to the unconstrained Fourier-neural-operator (FNO) model for the heat equation has a nonzero flux at the left insulating boundary, which means that it allows heat to flow through an insulator. This is in direct contradiction to the physics-enforced boundary constraint. BOON, which satisfies this so-called Neumann boundary condition, ensures that the gradient is zero at the insulator. Similarly, at the right boundary, we see that the FNO solution has a negative gradient at a positive heat source and that the BOON solution corrects this nonphysical result. Guaranteeing no violation of the underlying physics is critical to the practical adoption of these deep-learning models by practitioners in the field.
Stokes's second problem.png
Stokes’s second problem. This figure shows the velocity profile and corresponding absolute errors over time obtained by BOON (top). BOON improves the accuracy at the boundary, which, importantly, also improves accuracy on the interior of the domain compared to the unconstrained Fourier-neural-operator (FNO) model (bottom), where the errors at the boundary propagate inward over time.
Initial condition.png
2-D Navier-Stokes lid-driven cavity flow initial condition. The initial vorticity field (perpendicular to the screen), which is defined as the curl of the velocity field. At the initial time step, t = 0, the only nonzero component of the horizontal velocity is given by the top constant Dirichlet boundary condition, which drives the viscous incompressible flow at the later time steps. The other boundaries have the common no-slip Dirichlet boundary condition, which fixes the velocity to be zero at those locations.

Navier-Stokes lid-driven flow
2-D Navier-Stokes lid-driven cavity flow vorticity field. The vorticity field (perpendicular to the screen) within a square cavity filled with an incompressible fluid, which is induced by a fixed nonzero horizontal velocity prescribed by the Dirichlet boundary condition at the top boundary line for a 25-step (T=25) prediction until final time t = 2.
2-D Navier-Stokes lid-driven cavity flow relative error.
The L2 relative-error plots show significantly higher relative error over time for the data-driven Fourier neural operator (FNO) compared to that of our constrained BOON model on the Navier-Stokes lid-driven cavity flow problem for both a random test sample and the average over the test samples.

Acknowledgements: This work would have not been possible without the help of our coauthor Michael W. Mahoney, an Amazon Scholar; coauthors and PhD student interns Derek Hansen and Nadim Saad; and mentors Yuyang Wang and Margot Gerritsen.

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