How to produce factually accurate automatic text summaries

New metric can be calculated 55 times as quickly as its state-of-the-art predecessor, making it practical for model training.

Abstractive summarization is the automatic extraction and recombination of phrases from a text in order to summarize that text. Deep-learning-based abstractive-summarization systems are usually trained to maximize the overlap between the summaries they generate and sample summaries in their training data.

The trouble with this approach is that a summary that overlaps significantly with a target summary may recombine phrases in factually inaccurate manner. In the example below, which concerns an upcoming boxing match, the summarization model correctly concludes that “has a chink in his armor” summarizes an important aspect of the input text, but it applies it to the wrong boxer:

Klitschko example.png
Conventional metrics for training abstractive-summarization models don’t account for factual accuracy.

Although abstractive-summarization models have become very good at generating fluent, syntactically correct text, their frequent factual inaccuracy has severely hampered their adoption.

In a paper we presented at this year’s meeting of the Association for Computational Linguistics (ACL), we describe a new metric for measuring the performance of abstractive-summarization models, which accounts for factual accuracy. We also describe a methodology for using our metric to train abstractive-summarization models.

Our metric adopts the same general strategy as the earlier QAGS metric, but it’s 55 times as fast to apply, which makes it more practical for model training.

QAGS-QUALS-Image.png
Our new summary-scoring metric, QUALS (bottom), uses the same strategy as the earlier QAGS (top) but has a simpler architecture, enabling it to generate a score 55 times as quickly.
Credit: Glynis Condon

Using QAGS as an evaluation metric, we compared models trained using our approach to models trained using traditional metrics and methodologies, and we found that our approach improved on the best-performing previous models by 15% on one dataset and by 2% on another.

Scoring through question answering

QAGS (which stands for question answering and generation for summarization) uses a four-step procedure to score a text summary. First, it extracts names and noun phrases from the summary; these are potential answers to potential questions about the summary. 

Second, it feeds each extracted noun, together with the text of the summary, to a trained question generation model, which produces a question whose answer is the noun. Third, it feeds each of the generated questions to a trained question-answering model, once accompanied by the summary and once accompanied by the source text. 

QAGS-Image.cropped.png
QAGS requires the sequential application of three neural models: an answer extraction model, a question-answering model, and a question generation model.
Credit: Glynis Condon

The final score assesses the similarity between the answers based on the source text and the answers based on the summary. The intuition is that if both the summary and the source text cause the question-answering model to answer the questions in the same way, the summary is factually accurate. If they cause different answers, then the summary has probably garbled some facts.

By accounting for factual accuracy, QAGS offers a better assessment of summary quality than metrics based on phrasal overlap. But it requires the sequential application of three different deep-learning networks, which is inefficient.

QUALS

Our approach, which we call QUALS (for question answering with language model score for summarization), reduces the number of models to one, which makes it 55 times as fast as QAGS.

That one model is the joint question-and-answer generation (QAGen) model that members of our group presented at last year’s ACL. It takes a text as input and generates question-and-answer pairs pertaining to it.

QUALS-Image.cropped.png
QUALS requires a single neural model, a question-and-answer generation model.
Credit: Glynis Condon

The output of the QAGen model for a given input can be thought of as a huge tree, in which the nodes are words and each edge encodes the likelihood that a particular word will be followed by another word.

For a given summary, we search the resulting tree to produce 60 high-probability question-and-answer pairs. Our search algorithm ensures that we explore diverse paths through the tree, in order to generate a variety of candidate questions and answers. Then we throw out all the question-answer pairs whose answers are not sequences of words found in the summary.

Next, we feed the source text on which the summary is based to the QAGen model. We use the resulting tree to calculate the probabilities of the same question-answer pairs we extracted for the summary. When, for the source text, the probability of generating a particular question-answer pair is small compared to the probability for the summary, the QUALS will be low. Intuitively, the discrepancy suggests that the question-answer pair was plausible for the summary but not in the source text, indicating factual inconsistency.

QUALS scoring.png
Probabilities per token (words and other standalone symbols) of two different question-answer pairs, based on a summary (blue) and an input document (orange). The large probability differences for the answer in the right-hand example give it a much lower QUALS score (-2.615) than the right-hand example (-0.054).

Training methodology

The QUALS score gives us an efficiently computable measure of a summary’s factual accuracy, but using it to train a machine learning model is not straightforward. Differences in QUALS score can’t simply be back-propagated through the QAGen model to update the summarization model.

So in our paper, we propose contrastive learning as a method for using QUALS to train a summarization model. First, we train a summarization model using the standard approach, which uses maximum-likelihood estimation (MLE) to approximate a phrasal-overlap score.

Next, we use the trained model to generate new summaries for all the source texts in the training data and create two different groups of summaries. One group, S+, contains ground truth summaries that have high QUALS scores (indicating factually accurate summaries); the other, S- contains generated summaries that have low QUALS scores (indicating factually inaccurate summaries).

Finally, we retrain the summarization model, using a loss function that encourages it to generate summaries like those in S+ and discourages it from generating summaries like those in S-.

Evaluation

Sample summaries.png
Examples from the human-evaluation study, featuring input texts and summaries produced using both MLE and the ConSeq model, which is trained using QUALS.

As baselines for the evaluation of our approach, we used two models. One was trained using MLE in the standard way, to fine-tune a BART language model. For the other, we used our contrastive-learning methodology, but instead of using QUALS to evaluate summaries, we used an ensemble of three ROUGE metrics (ROUGE 1, ROUGE 2, and ROUGE L), all of which are based on phrasal overlap.

In addition to evaluating the models’ performance using QAGS, we evaluated them according to the three ROUGE metrics and FactCC, another model-based metric that simply predicts the factual consistency of two texts. On all five metrics, models trained using QUALS outperformed the two baselines.

For validation, we also conducted a human-evaluation study, which involved 100 summaries generated using QUALS and 100 summaries generated using MLE for each of two datasets (XSUM and CNNDM). Human subjects were asked to compare the summaries on three attributes: factual consistency, informativeness and grammatical correctness.

On average, annotators found the QUALS-based summaries more factually accurate and more informative than the MLE-based summaries, for both datasets. On grammatical correctness, the two models’ performance was virtually indistinguishable.

Human-study stats.png
The results of the human-evaluation study. Subjects were asked whether summaries produced using QUALS were better than, worse than, or equal to those produced using MLE, on three axes.

Research areas

Related content

US, CA, Sunnyvale
The Region Flexibility Engineering (RFE) team builds and leverages foundational infrastructure capabilities, tools, and datasets needed to support the rapid global expansion of Amazon's SOA infrastructure. Our team focuses on robust and scalable architecture patterns and engineering best practices, driving adoption of ever-evolving and AWS technologies. RFE is looking for a passionate, results-oriented, inventive Data Scientist to refine and execute experiments towards our grand vision, influence and implement technical solutions for regional placement automation, cross-region libraries, and tooling useful for teams across Amazon. As a Data Scientist in Region Flexibility, you will work to enable Amazon businesses to leverage new AWS regions and improve the efficiency and scale of our business. Our project spans across all of Amazon Stores, Digital and Others (SDO) Businesses and we work closely with AWS teams to advise them on SDO requirements. As innovators who embrace new technology, you will be empowered to choose the right highly scalable and available technology to solve complex problems and will directly influence product design. The end-state architecture will enable services to break region coupling while retaining the ability to keep critical business functions within a region. This architecture will improve customer latency through local affinity to compute resources and reduce the blast radius in case of region failures. We leverage off the sciences of data, information processing, machine learning, and generative AI to improve user experience, automation, service resilience, and operational efficiency. Key job responsibilities As an RFE Data Scientist, you will work closely with product and technical leaders throughout Amazon and will be responsible for influencing technical decisions and building data-driven automation capabilities in areas of development/modeling that you identify as critical future region flexibility offerings. You will identify both enablers and blockers of adoption for region flex, and build models to raise the bar in terms of understanding questions related to data set and service relationships and predict the impact of region changes and provide offerings to mitigate that impact. About the team The Regional Flexibility Engineering (RFE) organization supports the rapid global expansion of Amazon's infrastructure. Our projects support Amazon businesses like Stores, Alexa, Kindle, and Prime Video. We drive adoption of ever-evolving and AWS and non-AWS technologies, and work closely with AWS teams to improve AWS public offerings. Our organization focuses on robust and scalable solutions, simple to use, and delivered with engineering best practices. We leverage and build foundational infrastructure capabilities, tools, and datasets that enable Amazon teams to delight our customers. With millions of people using Amazon’s products every day, we appreciate the importance of making our solutions “just work”.
US, WA, Seattle
The Sponsored Products and Brands team at Amazon Ads is re-imagining the advertising landscape through generative AI technologies, revolutionizing how millions of customers discover products and engage with brands across Amazon.com and beyond. We are at the forefront of re-inventing advertising experiences, bridging human creativity with artificial intelligence to transform every aspect of the advertising lifecycle from ad creation and optimization to performance analysis and customer insights. We are a passionate group of innovators dedicated to developing responsible and intelligent AI technologies that balance the needs of advertisers, enhance the shopping experience, and strengthen the marketplace. If you're energized by solving complex challenges and pushing the boundaries of what's possible with AI, join us in shaping the future of advertising. We are looking for an Applied Scientist to build the science that helps Amazon's advertisers grow — with a focus on Cross Border Sellers who face distinct barriers as they scale across multiple marketplaces, and this role is about understanding those pain points deeply and removing them: building intelligent, autonomous solutions that simplify advertising, act efficiently on the advertiser's behalf, and let advertisers accelerate their growth and success. We expect Scientist to think innovatively about how to reduce the effort and complexity of advertising for these advertisers. Working backwards from their needs — spanning hands-off sellers and global brands with cross-marketplace operations — you will take the lead on medium-to-large, ambiguous problems where neither the problem nor the solution is well defined, invent new methods, validate them through rigorous experimentation, and deliver customer-facing products with measurable business impact. This role combines science depth, product focus, and hands-on engineering: you will raise the science bar, build consensus on approach across product and engineering partners, and mentor scientists and engineers while remaining deeply hands-on with the hardest technical problems. Key job responsibilities As an Applied Scientist on this team you will: - Understand the pain points of Cross border advertisers as they scale across marketplaces, and build science-driven solutions that remove barriers and accelerate their growth and success. - Build agentic and ML systems that autonomously create, structure, and manage ad campaigns on advertisers' behalf, encoding auction and marketplace dynamics (bidding, budget pacing, targeting decisions) while balancing advertiser ROI, shopper experience, and marketplace health. - Innovate on new, simpler ways to advertise powered by GenAI, and push the frontier of existing autonomous programs (e.g., auto-targeting, global lift-and-shift) while proposing and prototyping the next generation of campaign automation. - Develop models across the campaign lifecycle — opportunity discovery, ranking, ad-readiness and demand prediction, and bid/budget optimization — and apply the right approach for each problem, from classical ML to LLM/reasoning methods. - Define and curate the datasets and signals needed to train and evaluate these systems — advertiser and campaign data, cross-marketplace performance, auction and bid/budget signals, impressions, clicks, conversions, and search-term/keyword performance. - Own core parts of the agentic architecture — planning, tool use and integration (e.g., MCP), reasoning frameworks (e.g., ReAct, CoT/ToT), and model customization — and define evaluation and safety methodology so that automated decisions are reliable and trustworthy. - Stay deeply hands-on: write production-quality, critical-path code and build core components that take systems from prototype to launch on large-scale pipelines (Spark/EMR, Airflow) and online serving. - Raise the science bar: review designs and experiment plans, and communicate results and tradeoffs clearly to technical and business leaders. About the team Autonomous SP drives growth and simplifies advertising for Amazon's hands-off advertisers by creating and enhancing autonomous campaign solutions. We lead existing successful programs, including auto-targeting and global lift-and-shift, and continually innovate by building new, simpler campaign constructs powered by GenAI to act efficiently on advertisers' behalf. Cross-border Seller Experience (CBSX) focuses on developing targeted solutions for domestic and global advertisers with multi-marketplace operations. By understanding and addressing their unique Seller Central needs, we optimize for cross-marketplace efficiency, improved performance, and streamlined advertising experiences. Our team operates horizontally, delivering impactful solutions that benefit both hands-on and hands-off advertisers globally. Together, we sit within Sponsored Products and Brands, which is re-imagining advertising through the latest generative AI — building responsible, intelligent systems that balance the needs of advertisers, the shopping experience, and marketplace health.
US, WA, Seattle
MULTIPLE POSITIONS AVAILABLE Employer: AMAZON WEB SERVICES, INC. Offered Position: Data Scientist III Job Location: Seattle, Washington Job Number: AMZ10593723 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 $166,036/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, 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, Seattle
Amazon Ads is one of the fastest-growing advertising businesses in the world, connecting brands with hundreds of millions of customers across shopping, streaming, gaming, audio and the open web. Behind every ad that runs on Prime Video, Twitch, IMDb and thousands of external publisher apps and sites is a decisioning platform built by Amazon Ads Supply Tech. That platform makes X trillion ad decisions every day and supports $XB+ in annual publisher revenue across XX+ Amazon-owned publishers and X,000+ external publishers in streaming TV, audio, display, mobile and podcasts. Every one of those decisions is a science problem: which inventory qualifies for which buyers, how publisher commitments are paced and delivered, which ad wins among competing candidates, and what price maximizes long-term yield. Each must be solved in milliseconds, under strict latency and cost budgets, while balancing advertisers, publishers, viewers and Amazon. At this scale, a fraction-of-a-percent model improvement is worth $X millions. We are looking for a Principal Applied Scientist to set the science strategy for this platform. You will decide where machine learning creates the most value, bring intelligence into real-time runtime systems, and lead models from idea to production. Your work will directly shape three of Amazon Ads’ biggest priorities: growing Amazon-owned and third-party supply, enabling full-funnel advertising, and building agentic, AI-driven workflows for publishers and advertisers. As a principal, you will be a technical leader across the organization. You will raise the science bar, mentor senior scientists, and influence product and engineering leaders. You will work alongside excellent, driven scientists, engineers and product leaders who take on hard problems and invest in each other’s growth. Few places in the world operate at this scale. Key job responsibilities - Define the ML strategy for the ad decisioning platform. Set a multi-year science vision across pricing, yield optimization, forecasting, ad selection and commitment delivery, tied to measurable business outcomes. - Identify opportunities to deploy intelligence in runtime systems. Find where rules and heuristics in real-time serving paths can be replaced or improved by learned models, and partner with engineering to fit them within latency, compute and reliability budgets. - Build ML models that operate at X trillion decisions a day, drawing on approaches such as deep learning, reinforcement learning and bandits, causal inference, forecasting, and auction and mechanism design. - Evaluate ML models rigorously. Design offline, counterfactual and off-policy evaluations and online experiments. Define metrics that capture long-term value for advertisers, publishers and viewers, not just short-term lift. - Refine models for output quality and performance. Improve accuracy, calibration and robustness, cut serving latency and cost, and monitor models in production for drift and degradation. - Lead science across the organization. Set technical standards, review science designs and launches, and mentor and grow senior and principal-track scientists. - Push the frontier, including LLM and agentic approaches, and contribute to Amazon’s internal and external science community through papers, talks and patents. A day in the life Start the day reviewing off-policy evaluation results for a new pricing model that projects an X% yield lift, and deciding with your team whether it is ready for an online test. Next, you work with engineers on fitting a deeper model into a millisecond latency budget at X billion requests. At midday, you dig into why a forecasting model drifted during a live sports event, and sketch a fix. Later, you review a scientist’s paper draft and present the science roadmap to product leaders. Throughout, you work with brilliant, driven colleagues who sharpen your thinking and help you grow. About the team Ads Decisioning team brings together applied scientists and product managers to build the intelligence behind Amazon’s publisher business. We serve Amazon-owned publishers such as Prime Video, Twitch and IMDb, along with thousands of external publishers across streaming TV, audio and display. Our models decide, in milliseconds, which inventory qualifies for which buyers, how commitments are delivered, which ad wins and at what price, balancing advertisers, publishers, viewers and Amazon. We pair scientific rigor with a bias for production impact, learn from each other, and invest in each other’s growth. Come solve problems few teams anywhere get to work on.