Amazon Halo Rise advances the future of sleep

Built-in radar technology, deep domain adaptation for sleep stage classification, and low-latency incremental sleep tracking enable Halo Rise to deliver a seamless, no-contact way to help customers improve sleep.

The benefits of quality sleep are well documented, and sleep affects nearly every aspect of our physical and emotional well-being. Yet one in three adults doesn’t get enough sleep. Given Amazon’s expertise in machine learning and radar technology innovation, we wanted to invent a device that would help customers improve their sleep by looking holistically at the factors that contribute to a good night’s rest.

That’s why we’re excited to announce that Amazon has unveiled its first dedicated sleep device — Halo Rise, a combined bedside sleep tracker, wake-up light, and smart alarm. Powered by custom machine learning algorithms and a suite of built-in sensors, Halo Rise accurately determines users’ sleep stages and provides valuable insights that can be used to optimize their sleep, including information about their sleep environments. Halo Rise has no sensors to wear, batteries to charge, or apps to open. And since a good wake-up experience is core to good sleep, Halo Rise features a wake-up light and smart alarm, designed to help customers start the day feeling rested and alert.

Halo Rise in action
A built-in radar sensor uses ultralow-power radio signals to sense respiration and movement patterns and determine sleep stages.

Designing with customer trust as our foundation

Customer privacy and safety are foundational to Halo Rise, and that's evident in both the hardware design and the technologies used to power the experience. Halo Rise features neither a camera nor a microphone and instead relies on ambient radar technology and machine learning to accurately determine sleep stages: deep, light, REM (rapid eye movement), and awake.

The technology at the core of Halo Rise is a built-in radar sensor that safely emits and receives an ultralow-power radio signal. The sensor uses phase differences between reflected signals at different antennas to measure movement and distance. Through on-chip signal processing, Halo Rise produces a discrete waveform corresponding to the user’s respiration. The device cannot detect noise or visual identifiers associated with an individual user, such as body images.

Using built-in radar technology enables us to prioritize customer privacy while still delivering accurate measurements and useful results. Customers have the option to manually put Halo Rise into Standby mode, which turns off the device’s ability to detect someone’s presence or track sleep.

Halo Rise hardware design
Halo Rise features a suite of sensors to accurately track your sleep and measure your room’s temperature, humidity, and light levels. 

Intuitive and accurate experience

To design the sleep-tracking algorithm that powers Halo Rise, we thought about the most common bedtime behaviors and the ways in which customers and their families (pets included) might engage with the bedroom. This led us to innovate on five main technological fronts:

  • Presence detection: Halo Rise activates its sleep detection only when someone is in range of the sensor. Otherwise, the device remains in a monitoring mode, where no data is transmitted to the cloud.
  • Primary-user tracking: Halo Rise distinguishes the sleep of the primary user (the user closest to the device) from that of other people or pets in the same bed, even though the respiration signal cannot be associated with individual users.
  • Sleep intent detection: Halo Rise detects when the user first starts trying to sleep and distinguishes that attempt from other in-bed activities — such as reading or watching TV — to accurately measure the time it takes to fall asleep, an important indicator of sleep health.
  • Sleep stage classification: Halo Rise reliably correlates respiration-driven movement signals with sleep stages.
  • Smart-alarm integration: During the user’s alarm window, the Halo Rise smart alarm checks the user’s sleep stage every few minutes to detect light sleep, while also maximizing sleep duration.
Halo-Vienna-MM_Wave-Chart.png
A combination of breathing and movement patterns enables Halo Rise to determine the primary user for the sleep session and to measure that person’s sleep throughout the night.

Presence detection

Halo Rise has an easy setup process. To get started, a customer will place Halo Rise on their bedside table facing their chest and note in the Amazon Halo app what side of the bed they sleep on — and that’s it: Halo Rise is ready to go. The radar sensor detects motion within a 3-D geometric volume that fans out from the sensor, an area called the detection zone. Within this zone, the presence detection algorithm estimates the location of the bed and an “out-of-bed” area between the bed and the device.

On-chip algorithms detect the motion and location of respiration events within the detection zone. In both cases — motion and respiration — the algorithm evaluates the quality of the signals. On that basis, it computes a score indicating its confidence that the readings are reliable and a user is present. Only if the confidence score crosses a reliability threshold does Halo Rise begin streaming sensor data to the cloud, where it is processed by the primary-user-tracking algorithm.

Radar Fan.png
The Halo Rise detection zone is the region within which the radar sensor senses motion and location.

Primary-user tracking

We know that many of our customers share their beds, be it with other people or with pets, so our algorithms are designed to track the sleep of only the primary user. Halo Rise starts a sleep session after it detects someone’s presence within the detection zone for longer than five minutes. From there, the primary-user-tracking algorithm runs continuously in the background, sensing the closest user’s sleep stages. As long as the user sleeps on their side of the bed, and their partner sleeps on the other side, Halo Rise will track the primary user’s sleep quality irrespective of who comes to bed first and who leaves the bed last.

During the sleep session, Halo Rise dynamically monitors changes in the user’s distance from the sensor, the respiration signal quality, and abrupt changes in respiration patterns that indicate another person’s presence. These changes cause the algorithm to reassess whether it’s actually sensing the intended user and to ignore the data unrelated to the primary user. For instance, if the user gets into bed after their partner has already fallen asleep, or if they use the restroom in the middle of the night, Halo Rise detects that and adjusts the sleep results accordingly.

Sleep intent detection

Another big algorithmic challenge we faced was determining when a user is quietly sitting in bed reading their Kindle or watching TV rather than trying to fall asleep. The time it takes to fall asleep (also known as sleep latency) is an important indicator of sleep health. Too short of a time may result from sleep deprivation, while too long of a time may be due to difficulty winding down.

To address this problem, we used a combination of presence and primary-user tracking along with a machine-learning model trained and evaluated on tens of thousands of hours of sleep diaries to accurately identify when the user is trying to sleep. The model uses sensor data streamed from the device — including respiration, movement, and distance — to generate a sleep intent score. The score is then post-processed by a regularized change-point detection algorithm to determine when the user is trying to fall asleep or wake up.

Halo Rise Sleep Intent v2.png
A machine learning model trained on thousands of hours of sleep uses respiration, movement, and distance data to generate a sleep intent score.

Sleep stage classification

Wearable health trackers like Halo Band and Halo View use heart rate and motion signals to determine sleep stages during the night, but Halo Rise uses respiration. To learn how to reliably recognize those stages, we needed to develop new machine learning models.

We pretrained a deep-learning model to predict sleep stages using a rich and diverse clinical dataset that included tens of thousands of hours of sleep collected by academic and research sources. The research included sleep data measured using the clinical gold standard, polysomnography (PSG). PSG studies use a large array of sensors attached to the body to measure sleep, including respiratory inductance plethysmography (RIP) sensors, whose output is analogous to the respiration data measured by Halo Rise.

Pretraining the model to predict sleep stages from RIP sensors enabled it to develop meaningful representations of the relationship between respiration and sleep prior to additional training on radar datasets collected alongside PSG. To collect radar training data for the models, we partnered with sleep clinics to conduct thousands of hours of PSG studies. Ultimately, this enables our models to classify sleep stages using just a built-in radar in the comfort of a customer’s home.

Halo_hypnogram.png
In the morning, customers can access a sleep hypnogram that provides a detailed breakdown of time spent in each sleep stage throughout the night.

A smarter wake-up experience

When woken naturally during a light sleep stage, people are most likely to feel rested, refreshed, and ready to tackle the day. Consequently, Halo Rise features a wake-up light, which gently simulates the colors and gradual brightening of a sunrise, and a smart alarm. Customers can also set an audible smart alarm that’s integrated with our sleep stage classification algorithms, optimizing their wake experience. Ahead of their scheduled wake-up time, the audible smart alarm monitors their sleep stages and wakes them up at their ideal time for getting up. This combination of wake-up light and smart alarm is shown to increase cognitive and physical performance throughout the day.

The smart-alarm algorithms are trained around two factors: sensing when the user is in light sleep and maximizing the user’s sleep duration. For the first component, Halo Rise needs to continuously monitor sleep stages during the alarm window — the 30 minutes before a user’s scheduled alarm — to identify when the user has entered a light sleep stage, known as the “wake window.”

At this phase, our algorithms work to sense “wakeable events,” such as a change in motion or breathing. This requires incrementally computing sleep stages to trigger the alarm with low latency. Unlike many sleep algorithms, Halo Rise does not require data from the entirety of the sleep session to classify sleep stages, allowing predictions to be used directly for alarm triggers as data is streamed.

For the second component, the system’s models are trained to predict the latest moment to trigger the alarm during the wake window. This ensures that as the user drifts between sleep stages, they are getting those crucial minutes of additional sleep before the alarm goes off.

The Halo Rise wake-up light
Halo Rise identifies a “wake window” when the user is in light sleep, while also maximizing sleep duration before activating an audible smart alarm.

A solution you can trust

To evaluate our machine learning algorithms, we collected thousands of hours of sleep studies comparing Halo Rise to PSG for over a hundred sleepers, developed with input from leading sleep labs. While sleep studies are typically conducted in sleep labs, we performed in-home PSG studies at participants’ homes under supervision of registered PSG technologists to test the device in naturalistic settings.

We used three different registered PSG technologists to reliably annotate ground truth sleep stages per the American Academy of Sleep Medicine’s scoring rules. We then compared Halo Rise’s outputs to the ground truth sleep data across 14 different sleep metrics — including time asleep, time awake, time to fall asleep, and accuracy for every 30 seconds — following analysis guidelines from a standardized framework for sleep stage classification assessment. This evaluation was supplemented by thousands of sleep diaries from our beta trials, expanding our evaluation to a diverse population of adults to account for variations in preferred sleep postures, age, body shapes, and other background conditions.

What’s next?

As we look to invent new products that help our customers live better longer, Halo Rise is an important step in giving our customers greater agency over their health and well-being. By looking holistically at the end-to-end sleep experience — not just going to sleep but also getting up in the morning — Halo Rise unlocks an entirely new way for customers to understand and manage sleep. We’re excited to help them make sense of valuable sleep data, from the quality and quantity of their sleep to their room’s environment, and deliver actionable insights and resources to improve it in the future. Halo Rise is just getting started, and we are going to learn from our customers how this technology can continue to evolve and become even more personalized to better meet their needs.

Research areas

Related content

IN, TS, Hyderabad
Welcome to the Worldwide Returns & ReCommerce team (WWR&R) at Amazon.com. WWR&R is an agile, innovative organization dedicated to ‘making zero happen’ to benefit our customers, our company, and the environment. Our goal is to achieve the three zeroes: zero cost of returns, zero waste, and zero defects. We do this by developing products and driving truly innovative operational excellence to help customers keep what they buy, recover returned and damaged product value, keep thousands of tons of waste from landfills, and create the best customer returns experience in the world. We have an eye to the future – we create long-term value at Amazon by focusing not just on the bottom line, but on the planet. We are building the most sustainable re-use channel we can by driving multiple aspects of the Circular Economy for Amazon – Returns & ReCommerce. Amazon WWR&R is comprised of business, product, operational, program, software engineering and data teams that manage the life of a returned or damaged product from a customer to the warehouse and on to its next best use. Our work is broad and deep: we train machine learning models to automate routing and find signals to optimize re-use; we invent new channels to give products a second life; we develop highly respected product support to help customers love what they buy; we pilot smarter product evaluations; we work from the customer backward to find ways to make the return experience remarkably delightful and easy; and we do it all while scrutinizing our business with laser focus. You will help create everything from customer-facing and vendor-facing websites to the internal software and tools behind the reverse-logistics process. You can develop scalable, high-availability solutions to solve complex and broad business problems. We are a group that has fun at work while driving incredible customer, business, and environmental impact. We are backed by a strong leadership group dedicated to operational excellence that empowers a reasonable work-life balance. As an established, experienced team, we offer the scope and support needed for substantial career growth. Amazon is earth’s most customer-centric company and through WWR&R, the earth is our customer too. Come join us and innovate with the Amazon Worldwide Returns & ReCommerce team!
GB, MLN, Edinburgh
We’re looking for a Machine Learning Scientist in the Personalization team for our Edinburgh office experienced in generative AI and large models. You will be responsible for developing and disseminating customer-facing personalized recommendation models. This is a hands-on role with global impact working with a team of world-class engineers and scientists across the Edinburgh offices and wider organization. You will lead the design of machine learning models that scale to very large quantities of data, and serve high-scale low-latency recommendations to all customers worldwide. You will embody scientific rigor, designing and executing experiments to demonstrate the technical efficacy and business value of your methods. You will work alongside a science team to delight customers by aiding in recommendations relevancy, and raise the profile of Amazon as a global leader in machine learning and personalization. Successful candidates will have strong technical ability, focus on customers by applying a customer-first approach, excellent teamwork and communication skills, and a motivation to achieve results in a fast-paced environment. Our position offers exceptional opportunities for every candidate to grow their technical and non-technical skills. If you are selected, you have the opportunity to make a difference to our business by designing and building state of the art machine learning systems on big data, leveraging Amazon’s vast computing resources (AWS), working on exciting and challenging projects, and delivering meaningful results to customers world-wide. Key job responsibilities Develop machine learning algorithms for high-scale recommendations problems. Rapidly design, prototype and test many possible hypotheses in a high-ambiguity environment, making use of both quantitative analysis and business judgement. Collaborate with software engineers to integrate successful experimental results into large-scale, highly complex Amazon production systems capable of handling 100,000s of transactions per second at low latency. Report results in a manner which is both statistically rigorous and compellingly relevant, exemplifying good scientific practice in a business environment.
US, CA, Palo Alto
Amazon’s Advertising Technology team builds the technology infrastructure and ad serving systems to manage billions of advertising queries every day. The result is better quality advertising for publishers and more relevant ads for customers. In this organization you’ll experience the benefits of working in a dynamic, entrepreneurial environment, while leveraging the resources of Amazon.com (AMZN), one of the world's leading companies. Amazon Publisher Services (APS) helps publishers of all sizes and on all channels better monetize their content through effective advertising. APS unites publishers with advertisers across devices and media channels. We work with Amazon teams across the globe to solve complex problems for our customers. The end results are Amazon products that let publishers focus on what they do best - publishing. The APS Publisher Products Engineering team is responsible for building cloud-based advertising technology services that help Web, Mobile, Streaming TV broadcasters and Audio publishers grow their business. The engineering team focuses on unlocking our ad tech on the most impactful Desktop, mobile and Connected TV devices in the home, bringing real-time capabilities to this medium for the first time. As a successful Data Scientist in our team, · You are an analytical problem solver who enjoys diving into data, is excited about investigations and algorithms, and can credibly interface between technical teams and business stakeholders. You will collaborate directly with product managers, BIEs and our data infra team. · You will analyze large amounts of business data, automate and scale the analysis, and develop metrics (e.g., user recognition, ROAS, Share of Wallet) that will enable us to continually measure the impact of our initiatives and refine the product strategy. · Your analytical abilities, business understanding, and technical aptitude will be used to identify specific and actionable opportunities to solve existing business problems and look around corners for future opportunities. Your expertise in synthesizing and communicating insights and recommendations to audiences of varying levels of technical sophistication will enable you to answer specific business questions and innovate for the future. · You will have direct exposure to senior leadership as we communicate results and provide scientific guidance to the business. Major responsibilities include: · Utilizing code (Apache, Spark, Python, R, Scala, etc.) for analyzing data and building statistical models to solve specific business problems. · Collaborate with product, BIEs, software developers, and business leaders to define product requirements and provide analytical support · Build customer-facing reporting to provide insights and metrics which track system performance · Influence the product strategy directly through your analytical insights · Communicating verbally and in writing to business customers and leadership team with various levels of technical knowledge, educating them about our systems, as well as sharing insights and recommendations
US, WA, Seattle
Amazon Advertising operates at the intersection of eCommerce and advertising, and is investing heavily in building a world-class advertising business. We are defining and delivering a collection of self-service performance advertising products that drive discovery and sales. Our products are strategically important to our Retail and Marketplace businesses driving long-term growth. We deliver billions of ad impressions and millions of clicks daily and are breaking fresh ground to create world-class products to improve both shopper and advertiser experience. With a broad mandate to experiment and innovate, we grow at an unprecedented rate with a seemingly endless range of new opportunities. The Ad Response Prediction team in Sponsored Products organization build advanced deep-learning models, large-scale machine-learning pipelines, and real-time serving infra to match shoppers’ intent to relevant ads on all devices, for all contexts and in all marketplaces. Through precise estimation of shoppers’ interaction with ads and their long-term value, we aim to drive optimal ads allocation and pricing, and help to deliver a relevant, engaging and delightful ads experience to Amazon shoppers. As the business and the complexity of various new initiatives we take continues to grow, we are looking for talented Applied Scientists to join the team. Key job responsibilities As a Applied Scientist II, you will: * Conduct hands-on data analysis, build large-scale machine-learning models and pipelines * Work closely with software engineers on detailed requirements, technical designs and implementation of end-to-end solutions in production * Run regular A/B experiments, gather data, perform statistical analysis, and communicate the impact to senior management * Establish scalable, efficient, automated processes for large-scale data analysis, machine-learning model development, model validation and serving * Provide technical leadership, research new machine learning approaches to drive continued scientific innovation * Be a member of the Amazon-wide Machine Learning Community, participating in internal and external MeetUps, Hackathons and Conferences
US, WA, Seattle
Prime Video is a first-stop entertainment destination offering customers a vast collection of premium programming in one app available across thousands of devices. Prime members can customize their viewing experience and find their favorite movies, series, documentaries, and live sports – including Amazon MGM Studios-produced series and movies; licensed fan favorites; and programming from Prime Video add-on subscriptions such as Apple TV+, Max, Crunchyroll and MGM+. All customers, regardless of whether they have a Prime membership or not, can rent or buy titles via the Prime Video Store, and can enjoy even more content for free with ads. Are you interested in shaping the future of entertainment? Prime Video's technology teams are creating best-in-class digital video experience. As a Prime Video technologist, you’ll have end-to-end ownership of the product, user experience, design, and technology required to deliver state-of-the-art experiences for our customers. You’ll get to work on projects that are fast-paced, challenging, and varied. You’ll also be able to experiment with new possibilities, take risks, and collaborate with remarkable people. We’ll look for you to bring your diverse perspectives, ideas, and skill-sets to make Prime Video even better for our customers. With global opportunities for talented technologists, you can decide where a career Prime Video Tech takes you! In Prime Video READI, our mission is to automate infrastructure scaling and operational readiness. We are growing a team specialized in time series modeling, forecasting, and release safety. This team will invent and develop algorithms for forecasting multi-dimensional related time series. The team will develop forecasts on key business dimensions with optimization recommendations related to performance and efficiency opportunities across our global software environment. As a founding member of the core team, you will apply your deep coding, modeling and statistical knowledge to concrete problems that have broad cross-organizational, global, and technology impact. Your work will focus on retrieving, cleansing and preparing large scale datasets, training and evaluating models and deploying them to production where we continuously monitor and evaluate. You will work on large engineering efforts that solve significantly complex problems facing global customers. You will be trusted to operate with complete independence and are often assigned to focus on areas where the business and/or architectural strategy has not yet been defined. You must be equally comfortable digging in to business requirements as you are drilling into design with development teams and developing production ready learning models. You consistently bring strong, data-driven business and technical judgment to decisions. You will work with internal and external stakeholders, cross-functional partners, and end-users around the world at all levels. Our team makes a big impact because nothing is more important to us than delivering for our customers, continually earning their trust, and thinking long term. You are empowered to bring new technologies to your solutions. If you crave a sense of ownership, this is the place to be.
US, WA, Bellevue
mmPROS Surface Research Science seeks an exceptional Applied Scientist with expertise in optimization and machine learning to optimize Amazon's middle mile transportation network, the backbone of its logistics operations. Amazon's middle mile transportation network utilizes a fleet of semi-trucks, trains, and airplanes to transport millions of packages and other freight between warehouses, vendor facilities, and customers, on time and at low cost. The Surface Research Science team delivers innovation, models, algorithms, and other scientific solutions to efficiently plan and operate the middle mile surface (truck and rail) transportation network. The team focuses on large-scale problems in vehicle route planning, capacity procurement, network design, forecasting, and equipment re-balancing. Your role will be to build innovative optimization and machine learning models to improve driver routing and procurement efficiency. Your models will impact business decisions worth billions of dollars and improve the delivery experience for millions of customers. You will operate as part of a team of innovative, experienced scientists working on optimization and machine learning. You will work in close collaboration with partners across product, engineering, business intelligence, and operations. Key job responsibilities - Design and develop optimization and machine learning models to inform our hardest planning decisions. - Implement models and algorithms in Amazon's production software. - Lead and partner with product, engineering, and operations teams to drive modeling and technical design for complex business problems. - Lead complex modeling and data analyses to aid management in making key business decisions and set new policies. - Write documentation for scientific and business audiences. About the team This role is part of mmPROS Surface Research Science. Our mission is to build the most efficient and optimal transportation network on the planet, using our science and technology as our biggest advantage. We leverage technologies in optimization, operations research, and machine learning to grow our businesses and solve Amazon's unique logistical challenges. Scientists in the team work in close collaboration with each other and with partners across product, software engineering, business intelligence, and operations. They regularly interact with software engineering teams and business leadership.
US, WA, Seattle
Come be a part of a rapidly expanding $35 billion dollar global business. At Amazon Business, a fast-growing startup passionate about building solutions, we set out every day to innovate and disrupt the status quo. We stand at the intersection of tech & retail in the B2B space developing innovative purchasing and procurement solutions to help businesses and organizations thrive. At Amazon Business, we strive to be the most recognized and preferred strategic partner for smart business buying. Bring your insight, imagination and a healthy disregard for the impossible. Join us in building and celebrating the value of Amazon Business to buyers and sellers of all sizes and industries. Unlock your career potential. We are seeking an Applied Scientist who has a solid background in applied Machine Learning and Data Science, deep passion for building data-driven products, ability to formulate data insights and scientific vision, and has a proven track record of executing complex projects and delivering business impact. Key job responsibilities • Data driven insights to accelerate acquisition of new members. • Develop and implement personalized marketing strategies and campaigns tailored to individual customer preferences, behaviors, and demographics to enhance engagement and drive customer loyalty. • Develop, implement, and optimize marketing attribution models to accurately measure the impact of various marketing channels and campaigns, and create valuation frameworks to assess the ROI and contribution of each channel to overall business objectives. • Work with a group of both applied scientists and software engineers to deliver machine-learning and data science solutions to production. • Advance team's engineering craftsmanship and drive continued scientific innovation as a thought leader and practitioner. • Mentor talented members, provide technical and career development guidance to both scientists and engineers in the organization. About the team The Marketing Science team applies scientific methods and research techniques to enhance our understanding of AB consumer behavior, market trends, and the effectiveness of marketing strategies. Our goal is to develop and advance theories and models that can be used to make informed decisions in marketing and to provide insights into consumer decision-making processes. Additionally, we seek to identify and explore emerging trends and technologies in marketing, and to develop innovative approaches for addressing the challenges and opportunities in the field.
US, WA, Seattle
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, in partnership with Region Flexibility, is driving an SDO-wide program to diversify our use of AWS regions beyond DUB, IAD, and PDX regions. The objective of the Diversify AWS Region Usage (DARU) program is to mitigate the risk of capacity concentration by encouraging teams to design workloads that are region-flexible, utilize AWS automation such as Flexible Fleets to access multiple capacity pools, and optimize workload placement so SDO efficiently utilizes AWS. This is a strategic, highly visible, multi-year program which spans all Amazon business. CloudTune is looking for a Data Scientist to join our forecasting team and support DARU program. The team develops sophisticated algorithms that involve learning from large amounts of past data, such as actual sales, website traffic, merchandising activities, promotions, similar products and product attributes to forecast the demand for our compute infrastructure. These forecasts are used to determine the level of investment in capital expenditures, promotional activity, engineering efficiency projects and determining financial performance. As a Data Scientist CloudTune, you will work with other scientists, software engineers, data engineers, and product managers on a variety of important applied machine learning problems in the area of time series modeling. You will be an expert at communicating insights and recommendations to audiences of varying levels of technical sophistication. You will lead the design, implementation, and delivery of data science solutions for complex capacity planning problems. Key job responsibilities - Research and develop new methodologies for capacity demand forecasting. - Translate analytic insights into concrete, actionable recommendations for business or product improvement. Develop and present these as papers to senior stakeholders. - Given anecdotes about anomalies or generate automatic scripts to define anomalies, deep dive to explain why they happen, and identify fixes. - Drive scalable solutions for multi-year capacity demand forecasting horizons. - Play an integral role in developing a roadmap to expand and enhance demand forecasting for cloud compute resources. - Create and track accuracy and performance metrics (both technical and business metrics). - Create, enhance, and maintain technical documentation, and present to other scientists, engineers and business leaders.
US, CA, Sunnyvale
The Artificial General Intelligence (AGI) team is looking for a passionate, talented, and innovative applied scientist with a strong background in deep learning and speech processing techniques. You will have an enormous opportunity to impact the customer experience, design, architecture, and implementation of an industry-leading product used every day by people you know. As an Applied Scientist, you will leverage Amazon’s heterogeneous data sources and large-scale computing resources to develop novel machine learning algorithms to advance the state of the art in speech and audio processing. Key job responsibilities · Conduct applied research project(s) effectively and know when to ask for help and when to work independently · Engage with an experienced cross-disciplinary staff to conceive and design innovative solutions for consumer products. · Actively participate and contribute to research activities including publications and patents · Work closely with an internal inter-disciplinary team, and outside partners to drive key aspects of product definition, execution and test. · Be proactive, flexible and able to succeed within an open collaborative peer environment
US, CA, Santa Clara
AWS AI/ML is looking for world class scientists and engineers to work on foundation models, large-scale representation learning, and distributed learning methods and systems. At AWS AI/ML you will invent, implement, and deploy state of the art machine learning algorithms and systems. You will build prototypes and innovate on new representation learning solutions. You will interact closely with our customers and with the academic and research communities. You will be at the heart of a growing and exciting focus area for AWS and work with other acclaimed engineers and world famous scientists. Large-scale foundation models have been the powerhouse in many of the recent advancements in computer vision, natural language processing, automatic speech recognition, recommendation systems, and time series modeling. Developing such models requires not only skillful modeling in individual modalities, but also understanding of how to synergistically combine them, and how to scale the modeling methods to learn with huge models and on large datasets. Join us to work as an integral part of a team that has diverse experiences in this space. We actively work on these areas: Hardware-informed efficient model architecture, training objective and curriculum design Distributed training, accelerated optimization methods Continual learning, multi-task/meta learning Reasoning, interactive learning, reinforcement learning Robustness, privacy, model watermarking Model compression, distillation, pruning, sparsification, quantization A day in the life Why AWS? Amazon Web Services (AWS) is the world’s most comprehensive and broadly adopted cloud platform. We pioneered cloud computing and never stopped innovating — that’s why customers from the most successful start-ups to Global 500 companies trust our robust suite of products and services to power their businesses. Inclusive Team Culture Here at AWS, it’s in our nature to learn and be curious. Our employee-led affinity groups foster a culture of inclusion that empower us to be proud of our differences. Ongoing events and learning experiences, including our Conversations on Race and Ethnicity (CORE) and AmazeCon conferences, inspire us to never stop embracing our uniqueness. Mentorship & Career Growth We’re continuously raising our performance bar as we strive to become Earth’s Best Employer. That’s why you’ll find endless knowledge-sharing, mentorship and other career-advancing resources here to help you develop into a better-rounded professional. Work/Life Balance We value work-life harmony. Achieving success at work should never come at the expense of sacrifices at home, which is why we strive for flexibility as part of our working culture. When we feel supported in the workplace and at home, there’s nothing we can’t achieve in the cloud.