Careers

Alexa for Shopping (Rufus) is Amazon's new AI-powered shopping assistant that combines the capabilities of Rufus and Alexa+ to provide a more personalized and intelligent shopping experience. We are building the future of AI-powered commerce, where every customer interaction is conversational, personalized, and proactive. We are searching for pioneers who are passionate about technology, innovation, and customer experience, and are ready to make a lasting impact on the industry. In this role, you will be managing a team working on Large Language Model (LLM) and/or Vision-Language Model (VLM) post-training and alignment for new shopping experiences. You will set the roadmap and vision for science investments across multiple new customer experiences and to optimize existing experiences to deliver the most helpful, accurate, and fastest AI shopping assistant in the industry The ideal candidate is deeply steeped in LLM-based architectures, post-training techniques (RLHF, DPO, fine-tuning), and multi-agent systems. They are passionate about applied science, working back from customer experience to define what matters, and building teams that ship production AI at scale. You’ll be working with talented scientists, engineers, and product managers to innovate on behalf of our customers. If you’re fired up about being part of a dynamic, driven team, then this is your moment to join us on this exciting journey!

Amazon's eCommerce Foundation (eCF) organization provides the core technologies that drive and power Amazon's Stores, Digital, and Other (SDO) businesses. Millions of customer page views and orders per day are enabled by the systems eCF builds from the ground up. CloudTune, within eCF, empowers growth and business agility needs by automatically and efficiently managing AWS capacity and business processes needed to safely meet Amazon's customer demand. CloudTune serves its primary customers, internal software teams, through forecast-driven automation of cost controllership, capacity management, and scaling. We predict expected load, and drive procurement and allocation of AWS capacity for new product launches and high-velocity events like Prime Day and Cyber Monday. CloudTune is looking for a Research Scientist to join our forecasting and optimization team. The team develops sophisticated algorithms that combine machine learning-based demand forecasting with mathematical optimization to solve large-scale capacity planning problems under uncertainty. We work with massive datasets — spanning thousands of availability zone and instance family combinations across global regions — to determine optimal resource allocation strategies that balance infrastructure holding costs against service availability requirements. These models directly inform multi-million dollar capacity investment decisions and drive automated procurement, retention, and release policies across Amazon's compute infrastructure. As a Research Scientist in CloudTune, you will work with other scientists, software engineers, data engineers, and product managers on a variety of important research problems in the areas of stochastic optimization, time series modeling, and operations research. You will formulate capacity planning challenges as constrained optimization problems, develop demand forecasting models that account for asymmetric risk, and design allocation frameworks that minimize costs while maintaining fulfillment guarantees. You will analyze and process large amounts of data, develop new algorithms and improve existing approaches based on statistical models, machine learning algorithms, and big data solutions to automatically scale Amazon's compute infrastructure, optimizing the balance between availability risk and cost efficiency for all of Amazon's businesses. Key job responsibilities - Formulate capacity planning and resource allocation challenges as mathematical optimization problems (linear programming, stochastic optimization, mixed-integer programming) - Develop demand forecasting models using machine learning (XGBoost, quantile regression, deep learning) with risk-aware loss functions tailored to operational objectives - Design and implement safety stock and retention policy optimization frameworks that balance holding costs against fulfillment risk across constrained and unconstrained capacity pools - Process and analyze large-scale operational data (order histories, capacity utilization, availability constraints) to identify patterns and inform model development - Create, enhance, and maintain technical documentation, and present research findings to scientists, engineering teams, and senior leadership

Alexa Connections is on a mission to become the world's most trusted communication agent, spanning calls, text, email, and the ever-expanding surfaces where people connect. We're building intelligence that keeps customers connected effortlessly while putting their trust and privacy first. As an Applied Scientist, you'll help build the smartest communications agent for people, an agent that understands intent, context, and relationships, and that acts on a customer's behalf to make every interaction feel effortless and genuinely helpful Key job responsibilities You'll research, develop, and deploy the language intelligence that powers how people communicate, using and adapting large language models to understand intent, context, and the entities that matter in a conversation, the people, contacts, events, dates, and topics that tell the agent who and what a message is about. In practice, that means reframing classic NLP and named entity recognition as LLM-native tasks through instruction-tuning, in-context learning, and structured generation, and building agentic capabilities that reason over multi-turn conversations and act on them, from summarization and smart replies to coreference and context resolution grounded in a customer's relationships and history. You'll fine-tune, align, and optimize foundation models for the messiness of real communication data, informal text, code-switching, misspellings, transcribed speech, and multilingual content, and ground their outputs in customer context using retrieval-augmented generation and personalization so responses stay relevant and reliable. Throughout, you'll take models from research to production at scale, partnering with engineering and product teams to ship LLM-powered features used by millions every day, and define the evaluation frameworks that measure quality, faithfulness, latency, and customer impact while keeping trust and privacy first.

Are You Ready to Redefine How the World Receives Its Packages? What if your algorithms defined the most efficient path for millions of deliveries — every single day? At Amazon, we're building the science that makes that possible, and we're looking for exceptional scientists to help lead the way. The Last Mile Routing & Planning organization develops the software, algorithms, and tools that power the "magic" of home delivery. Our planning and routing intelligence systems drive billions of daily decisions — enabling safe, efficient, and frustration-free routes for drivers across the globe. What You'll Do In this role, you'll sit at the intersection of state-of-the-art research and real-world impact. You will: - Design and build algorithms that solve large-scale, complex logistics problems - Synthesize data from diverse sources to identify high-value business opportunities - Provide research direction and data-driven insights to guide strategic decisions - Translate complex technical approaches into clear communication for scientists, engineers, and business stakeholders - Partner closely with scientists and engineers in a collaborative, high-impact environment What You'll Work On We have an exciting and growing portfolio of research areas, including: - Routing for same-day and grocery deliveries - Planning for electric and autonomous vehicles - District-level and stop-level planning - Forecasting solutions for diverse delivery programs All of this is powered by the latest methods in Operations Research (OR), Machine Learning (ML), and Generative AI — at a truly global scale. Successful candidates will lead one or more of these problem spaces. What We're Looking For - Deep expertise in Operations Research and/or Machine Learning methods - Proven experience applying these methods to large-scale, real-world business problems - Ability to translate models into production-ready code in Python or Java - Strong communication skills — you can explain complex technical concepts to diverse audiences - A bias for action and an iterative mindset when tackling ambitious research challenges Why Amazon We're passionate about your growth. Whether you want to explore emerging technologies, take on broader scope, or accelerate your career trajectory, we'll invest in helping you get there. Our business is scaling fast — and so are the opportunities for the people who build it. If you're driven by the challenge of optimizing one of the world's most complex logistics systems and excited to see your work impact millions of customers daily, we'd love to hear from you. Key job responsibilities - Invent and design novel solutions for scientifically complex problem areas, and identify opportunities for invention within existing and new business initiatives - Deliver large-scale, high-impact solutions to complex problems in support of medium-to-large business goals - Shape the design of scientifically complex software systems, personally contributing significant portions of the critical scientific novelty - Apply mathematical optimization, machine learning, and Generative AI techniques to develop solution methodologies for in-house decision support tools and software - Research, prototype, simulate, and experiment with models — and actively participate in their production-level deployment in Python or Java - Engage with the broader scientific community by publishing research articles and participating in leading research conferences

We are looking for a Senior Inference Engineer to own inference for real-time multimodal conversational AI. This is a full-stack inference role: you will work across the entire path a model takes from research to production — shaping model architecture so it is servable, building the real-time runtime that serves it within hard latency budgets, and building the offline systems that train and reinforce it. You will operate at the boundary of Science and Inference, taking frontier-scale speech and audio models and making them run within real-time latency budgets on production hardware. You will co-design architectures with scientists to make them inference-friendly from inception, own the low-latency streaming serving path, and build the training and reinforcement-learning infrastructure that closes the loop. You will have the compute, data, and runway to solve problems that few teams in the world are positioned to tackle. As a Senior Engineer, you will own a significant area of the inference stack end to end, drive its technical execution, contribute to the team's roadmap, and work closely with scientists and hardware partners to ensure our models run fast enough to feel human in real time — and at a cost that makes them viable at scale. You may go deep in one of the areas below while contributing across the others. Key job responsibilities Model Architecture & Inference Co-Design • Partner with research scientists to make model architectures servable from inception — surfacing the latency, memory, and cost implications of architecture choices before they are locked in • Implement and optimize the inference path for large-scale multimodal models — attention and KV-cache mechanisms, multimodal/autoregressive decoding, and the compute primitives on the critical path Apply efficiency techniques across the stack — quantization (per-tensor/per-channel/per- group, INT8/FP8/BF16), speculative decoding, operator fusion, and paged KV-cache — and quantify their quality/latency trade-offs • Develop and tune high-performance kernels for critical operations where off-the-shelf implementations leave performance on the table, integrating them into production serving with minimal overhead • Profile end-to-end performance with tools such as Nsight Compute/Systems and roofline analysis to identify and eliminate bottlenecks in large-scale inference workloads Real-Time & Interactive Runtime • Own the real-time serving path for streaming multimodal conversational AI, meeting sub- second, streaming latency budgets under concurrent session load • Build and tune continuous batching, scheduling, and preemption to balance throughput against per-request latency SLAs for interactive workloads • Customize production serving frameworks (e.g., vLLM, PyTorch) for real-time streaming generative models that fall outside standard LLM serving patterns — sustained low-latency output under concurrent session load • Implement multi-GPU inference (tensor parallelism, collective communication) for latency- critical paths, and drive cost toward parity with existing production baselines • Establish latency, throughput, and cost benchmarking, and publish the operational metrics that gate deployment Offline Systems: Training, RL & Evaluation Infrastructure • Build and scale the offline inference systems behind post-training — high-throughput rollout generation and reward-model serving for reinforcement learning (RL/RLHF/RLAIF) • Ensure train/serve consistency — that the inference path used in RL and evaluation faithfully matches production online behavior (e.g., parity across sampling and logit processing) • Work with the evaluation team to enable offline inference that captures the quality dimensions unique to real-time conversation — latency sensitivity, audio quality, and interaction naturalness

Have you ever placed an order on Amazon and wondered how it got to you so fast? Behind that speed is a massive transportation network generating billions of data points daily. We need someone who can turn that data into clarity. Come join the Network Engineering, Scheduling and Technology (NEST) Science team within Amazon Transportation Services. We are looking for a Data Scientist who is equal parts data engineer, visualization architect, and analytical modeler. You will own the end-to-end build process for data-driven solutions: identifying business needs, developing simulation and optimization models, building computationally efficient analytical tools, and narrating results through compelling data storytelling. This is not a dashboard-building role. You will work at the intersection of large-scale data processing, advanced analytics (including simulation and optimization), and data visualization, building tools that allow stakeholders to explore millions of records interactively, uncover patterns in network performance, and make data-driven decisions with confidence. The ideal candidate is a data wizard who thrives on wrangling massive datasets, building predictive and prescriptive models, architecting performant query and aggregation pipelines, and crafting visualizations that communicate complex findings with precision and clarity. You will own the full lifecycle, from problem identification and data extraction through modeling and simulation to production-grade analytical applications that narrate results back to stakeholders. You will collaborate closely with scientists, engineers, and product managers Key job responsibilities - Own the end-to-end analytical lifecycle: identify stakeholder needs, frame problems, build models, and narrate results through data tools and visualizations - Design and build production-grade analytical tools, BI applications, and interactive data products that enable self-service exploration of very large transportation datasets (billions of records) - Develop and enhance simulation and optimization models (discrete event simulation, agent-based modeling, mathematical optimization) applied to network planning and transportation operations - Architect computationally efficient data pipelines and aggregation strategies that support responsive, real-time or near-real-time visualization at scale - Develop advanced data storytelling artifacts that communicate complex network dynamics, trends, and anomalies to technical and non-technical stakeholders - Build and maintain reusable visualization frameworks and libraries tailored to transportation network data (routing, scheduling, flow, capacity) - Work with large-scale data platforms (Redshift, Spark, S3, Athena) to extract, transform, and model data for analytical consumption - Develop code (Python, SQL, Scala) for data processing, statistical modeling, simulation, and building automated analytical workflows - Collaborate with Applied Scientists, Research Scientists, Software Engineers, and Product Managers to integrate analytical tools into broader planning and decision-support systems - Define and implement best practices for data visualization performance, including sampling strategies, level-of-detail rendering, and progressive loading for large datasets - Communicate findings, methodology, and recommendations through compelling written and verbal presentations to leadership and business customers About the team The Network Engineering, Scheduling, and Technology (NEST) Science Team prototype, build, and productionize mathematical models that reduce transportation cost and improve customer experience in Amazon's Middle Mile network. Equipped with techniques from Operations Research, Machine Learning and Simulation, these models are used to govern scheduling and equipment selection of hundreds of thousands of truck movements, optimize network configurations, determine the transit times between nodes, and simulate network flow under uncertainty for informed decision making. Our core team consists of Applied, Data, and Research Scientists along with technical Product Managers that come from diverse backgrounds.

Amazon serves hundreds of millions of customers. Each one has a unique history of purchases, preferences, and behaviors. Our team's mission: turn that history into real-time contextual intelligence that makes every Amazon experience feel personal. We're hiring an Applied Scientist to push the boundaries of what's possible with LLMs, semantic retrieval, and customer understanding at scale. The problem space: Imagine a system that can instantly synthesize years of customer signals — what they bought, what they love, what they're planning — and surface the exact right context for any experience, in milliseconds. That's what we build. It's equal parts information retrieval, generative AI, and systems engineering. Why this role: 1. Scale: Your models will serve 1,500+ requests per second across Amazon's largest surfaces. 2. Impact: Direct revenue attribution in the hundreds of millions — your work shows up in customer experiences the same week. 3. Frontier tech: Fine-tuning LLMs, building custom embedding models, designing retrieval architectures that balance quality with sub-100ms latency constraints. 4. Data richness: Access to one of the most comprehensive customer behavior datasets anywhere. 5. Ownership: End-to-end — from research to production deployment to metric evaluation. Key job responsibilities 1. Invent new approaches to contextual retrieval, relevance scoring, and LLM-based summarization. 2. Fine-tune and evaluate language models for domain-specific understanding. 3. Design experiments that measure real customer impact, not just benchmark scores. 4. Ship production systems and iterate based on live metrics. 5. Collaborate across teams — Alexa, Search, Recommendations — as a platform that powers them all. 6. Mentor team members and shape the technical direction of our roadmap. Please visit https://www.amazon.science for more information. A day in the life You'll analyze large-scale behavioral data, design experiments, and build models that ship to production. You'll work closely with engineers to ensure your science translates into low-latency, high-reliability systems. You'll present findings to leadership and influence product strategy. Some weeks you'll be deep in model architecture; other weeks you'll be debugging a relevance gap in production. Every day, your work reaches real customers. About the team We're a small, high-impact team that values scientific rigor and engineering craft equally. The team values innovations and offers a safe place to try, fail and learn while fostering a culture of continuous improvement. Everyone is a leader and owner for everything we do as a team. Our team offers creative space with entrepreneurial work environment focusing on customer obsession.

Amazon's Worldwide Grocery Stores (WWGS), Data & Science team is seeking an Applied Scientist to join our Sales & Operations Planning (S&OP) and Supply Chain Science team. In this role, you will help build machine learning models that improve how the Amazon Grocery Network plans and stocks its stores, where gaps between plan and reality lead directly to out-of-stocks, wasted product, higher costs, and degraded customer experience. You will contribute to the development and deployment of models across a range of grocery supply chain problems, including demand forecasting, customer preference modeling, and improving product availability, using time series, Bayesian and structural methods, and machine learning. You will work alongside senior scientists who will help you scope problems, review your designs and code, and grow your depth in supply chain science and production ML — and you will work closely with engineering partners, product owners, and business stakeholders to deliver measurable impact. Our models inform planning and inventory decisions across the grocery supply chain, many of them carried out by partner teams and the systems they own, so understanding how model errors land on stores, planners, and customers matters as much as improving offline metrics. You will participate in design and roadmap discussions, communicate clearly with technical and non-technical partners, and develop judgment about the trade-offs in the systems you contribute to. We are investing in Generative AI to advance supply chain workflows, moving from human-in-the-loop to AI-in-the-loop decision support. Opportunities include automating routine planner interventions, surfacing recurring sources of operational defects, and augmenting planner and scientist judgment with agentic tools. Key job responsibilities - Develop, evaluate, and deploy components of machine learning and statistical models for grocery supply chain problems, including demand forecasting, customer preference modeling, and product availability, with input and guidance from senior scientists. - Build models and mechanisms that reduce out-of-stocks and shrink, including identifying and helping correct upstream data and process issues that degrade them. - Translate business problems into well-defined scientific solutions with clear objectives, constraints, and success metrics, partnering with senior scientists on the more ambiguous ones. - Analyze model performance and downstream impact on inventory, availability, and capacity decisions; contribute to metrics that reflect business outcomes, not only offline model accuracy. - Prototype and evaluate Generative AI approaches in our supply chain workflows, including automated interventions, and help productionize the ones that prove out. - Partner with engineering teams to productionize models, contribute to data pipelines, and build scalable, maintainable science systems. - Monitor deployed models, investigate performance issues, and continuously improve model quality and calibration. - Communicate technical concepts and recommendations clearly through documentation, presentations, and design reviews with scientists, engineers, product managers, and business leaders. - Contribute to the internal scientific community through knowledge sharing and, where appropriate, research publications.

Amazon is looking for an Applied Scientist to help build next generation selection/assortment systems. On the Specialized Selection team within the Supply Chain Optimization Technologies (SCOT) organization, we own the selection of the products that Amazon offers in our limited shelf assortment problems world wide. This includes products for our fastest delivery, perishable grocery offerings, and other emerging Amazon delivery programs. The selection is generated with a series of Machine Learning (ML) and optimization models to best cater to customer purchase intents under limited warehouse capacity. We build tools and systems that enable our partners and business owners to scale themselves by leveraging our problem domain expertise, focusing instead on introspecting our outputs and iteratively helping us improve our models rather than hand-managing their assortment. We partner closely with our business stakeholders as we work to develop state-of-the-art, scalable, automated selection management systems. As an Applied Scientist, you will work with software engineers, product managers, and business teams to understand the business problems and requirements, distill that understanding to crisply define the problem, and design and develop innovative solutions to address them. Our team is highly cross-functional and employs a wide array of scientific tools and techniques to solve key challenges, including supervised and unsupervised machine learning, large language models, mixed integer linear programs (MILPs), reinforcement learning, causal inference, and experiment designs. Some critical research areas in our space include modeling substitutability between similar products, complementarity and basket building, measuring speed sensitivity of products through experiments, optimizing assortment under operational and capacity constraints, and supply and demand forecasting. Key job responsibilities You will be an end-to-end owner for the projects you support. Responsibilities include: Understanding business requirements and existing challenges and map them to the right scientific solution; Designing effective, scalable, and achievable solutions to key business problems; Developing the right set of metrics to evaluate efficacy of your models and solutions; Prototyping and analyzing new models and business logic; Productionizing your scientific solutions, including writing production-quality critical path code; Communicating, both written and verbally, with both technical and business audiences throughout each project; Publishing findings in internal and/or external conferences and interfacing with the scientific community; Mentoring and developing the scientist community across the organization

Are you passionate about solving complex problems and protecting one of the world’s largest cloud platforms? The AWS Payments and Fraud Prevention team is looking for an innovative Applied Scientist to help keep AWS a safe and trusted environment for millions of customers worldwide. In this role, you will design, build, and deploy machine learning models that detect, prevent, and mitigate fraudulent activity across the AWS ecosystem. You will work with massive, real-world datasets, develop new detection strategies, and apply advanced and practical technologies to tackle ever-evolving threats. You will also explore Generative AI (GenAI) techniques to uncover new fraud patterns and strengthen our fraud defenses. At AWS, we support hundreds of thousands of businesses, powering billions of transactions every day. Fraudsters are constantly innovating — and so are we. If you enjoy thinking like a fraudster, building resilient defenses, and making a real-world impact, we invite you to join us and help shape the future of secure cloud computing. Key job responsibilities * Design, build, and deploy machine learning models to detect, prevent, and mitigate fraudulent activities across the AWS platform. * Analyze large-scale behavioral, transactional, and historical datasets to uncover fraud patterns and emerging threats. * Explore and apply GenAI techniques, including large language models (LLMs), synthetic data generation, and adversarial simulations to enhance fraud detection capabilities. * Collaborate closely with engineering, product, and operations teams to translate business needs into scalable technical solutions. * Experiment, prototype, and iterate on new detection strategies, algorithms, and evaluation metrics. * Continuously monitor model performance and improve robustness against adversarial behaviors and evolving fraud tactics. * Communicate findings and technical insights clearly and effectively to both technical and non-technical audiences. * Contribute to the broader fraud prevention strategy, driving innovation and best practices across the organization. About the team AWS Payments and Fraud Prevent is responsible for detecting & mitigating AWS account risks. You’ll be part of a team of Scientists, Analysts, and Technical & non-Technical Program Managers. The team’s goal is to identify and neutralize fraudsters from unauthorized access to legitimate AWS customers accounts. We have a formal mentor search application that lets you find a mentor that works best for you. Your manager can also help you find a mentor or two, because two is better than one. In addition to formal mentors, we work and train together so that we are always learning from one another, and we celebrate and support the career progression of our team members.