The science behind Echo Frames

How the team behind Echo Frames delivered longer battery life and improved sound quality inside the slim form factor of a pair of eyeglasses.

When the team behind Amazon’s Echo Frames set out to improve the next generation of their product, they needed to strike a delicate balance. Customer feedback on earlier versions of the smart audio eyeglasses centered on three elements: longer battery life, more style options, and improved sound quality.

A man with a beard is seen wearing a pair of Echo Frames glasses. He is standing outside and is pictured in three-quarters view.
Echo Frames feature custom-built speech processing technology that drastically improves word recognition — key for interacting with Alexa in windy or noisy environments.

Achieving all three of those goals would be a challenge in itself; doing that inside the slim form factor of a pair of Alexa-enabled eyeglasses upped the ante.

“All three of those goals are in tension with one another,” says Adam Slaboski, senior manager of product management and product lead for Echo Frames. The easiest way to improve battery and audio would be to increase the size of the device, but that would conflict with feedback around the importance of design. Amping up bass to improve the audio experience would consume more battery, and so on.

Finding that sweet spot was a huge effort in engineering and customer understanding.
Adam Slaboski

“Finding that sweet spot was a huge effort in engineering and customer understanding," Slaboski says.

With Echo Frames (3rd Gen) and Carrera Smart Glasses with Alexa (designed in collaboration with Safilo, one of the world’s leading eyewear companies), the Smart Eyewear team met the challenge.

The smart glasses feature enhanced audio playback, with custom-built speech-processing technology that dramatically improves word recognition — key for interacting with Alexa in windy or noisy environments. The new range of frame styles come in a variety of sizes, and all come with a significant boost in battery life.

From the outside, Echo Frames still look like a pair of regular eyeglasses. “But we changed everything on the inside,” says Jean Wang, general manager and director of Smart Eyewear. “And we learned new lessons along the way.”

Here’s how Amazon engineers and product designers tackled all three customer demands.

Turning up the volume with open-ear audio

Like previous generations of Echo Frames, the current model uses open-ear audio. In addition to fitting the form factor of a pair of glasses, this allows users to maintain awareness of their surroundings while interacting with Alexa or enjoying audio entertainment.

Related content
Combining psychoacoustics, signal processing, and speaker beamforming enhances stereo audio and delivers an immersive sound experience for customers.

The open-ear audio design has been popular with users who are blind or have low vision, notes Jenai Akina, senior product manager for Echo Frames. “It’s really beneficial that it doesn’t obstruct a critical sense like hearing,” she explains. “That form factor is really helpful for daily interactions — especially when we want to be open to engage with our environment and the people around us. Open ear allows customers to maintain awareness, while providing access to a voice assistant.”

Open-ear audio brings a host of unique challenges to the engineering process. Typical headphones and earbuds block off the ear from the outside world, preventing air from escaping. That funnels more of the sound waves from the speakers into the user’s ears. With an open-ear design, sound has to travel farther, and there is less control over direction. That could lower the audio volume and reduce clarity — and importantly, audio could leak out to people standing nearby. The key is to drive the sound pressure as much as possible toward the user’s ears while minimizing the audio leakage.

By bringing people into the lab, we can simulate real environmental noise conditions like wind, background noise in a crowded restaurant, and the sound of cars on the road.
Scott Choi

In working to improve audio quality, the team continued to hone the directionality of the sound while also working to improve volume and bass. A technique called dipole speaker configuration helps to do both. In addition to a sound porthole located near the ear canal, the frames feature a second porthole that cancels unnecessary sound while amping up bass.

With input from in-house audio experts and instruments to analyze measurements like harmonic distortion, the team came up with a set of potential tuning solutions that met objective targets for audio quality. They then tested those “flavors” of tuning in the lab with several user groups.

“By bringing people into the lab, we can simulate real environmental-noise conditions like wind, background noise in a crowded restaurant, and the sound of cars on the road,” explains senior manager of audio Scott Choi. That allowed his team to understand environmental variables in a controlled setting.

With the feedback from those focus groups, the team then selected a few of the most popular tunings to push out to beta testing, where users could provide feedback on a weekly basis.

“We see how the feedback trends change with each tuning change, which gradually allows it to mature and converge into a certain tuning,” Choi says. The result is audio calibrated to maximize intelligibility and volume without leaking private conversations (or guilty-pleasure playlists).

The Echo Frame team used a rotating arch of microphones to lest leakage. This animation shows the array moving in circles around a mannequin wearing the Gen 3 prototype, creating a 3D sphere plot of audio leakage. Via this testing, the team was able to minimize leakage to the side and back.
The Echo Frame team used a rotating arch of microphones to lest leakage. The array moved in circles around a mannequin wearing the Gen 3 prototype, creating a 3D sphere plot of audio leakage. Via this testing, the team was able to minimize leakage to the side and back.

To test leakage, the audio team rigged up a rotating arch of microphones. The array moved in circles around a mannequin wearing the Gen 3 prototype, creating a 3-D sphere plot of audio leakage. Choi explains that they focused on minimizing leakage to the side and back, and ultimately, the speakers were moved much closer to the ear to help minimize leakage and improve loudness.

Leakage isn’t the only privacy consideration. The Echo Frame team also continues to innovate on protecting users from bad actors who may get hold of their smart glasses.

Related content
Amazon senior principal engineer Luu Tran is helping the Alexa team innovate by collaborating closely with scientist colleagues.

Gen 2 protected users by requiring them to authenticate their sessions using a trusted phone. Without authentication, a user can’t invoke sensitive commands like “navigate me home,” unlocking a smart lock, or making a purchase. But customers didn’t like the added friction.

Now customers who enroll in Alexa Voice ID will be able to use their vocal fingerprints for authentication to receive responses to smart-home utterances.

“We’re the first on-the-go Alexa device to use Voice ID for privacy authentication,” Slaboski says.

Boosting battery life without cramping style

Gen 3 improves continuous music playback time to six hours, versus the four hours offered by the previous generation of Echo Frames. It also bumps battery life to up to 14 hours of moderate usage spread across playback, talk time, notifications, and Alexa interactions.

Delivering the desired loudness, bass, and audio quality while optimizing for battery life was a careful balance.
Ravi Sanapala

The team couldn’t simply slap on a bigger battery without making the Echo Frames look less like normal glasses. And with sound quality high on the priority list as well, the devices were going to need as much juice as ever. The team focused on trimming power use in standby mode, ensuring that the overall battery consumption would go down without weakening the speakers when users needed them.

“Delivering the desired loudness, bass, and audio quality while optimizing for battery life was a careful balance,” says senior product manager Ravi Sanapala. “We need the battery to last throughout as much of the day as possible and for Alexa to be available whenever users need it.”

The architectural changes in speaker placement helped keep power needs low while improving audio. The team also tweaked the placement of the battery itself, distributing its capacity differently than in Gen 2. Sanapala adds that algorithmic changes were key in balancing idle-battery conservation and on-demand device usage.

“We had to collaborate with all of our cross-functional teams to optimize everything,” Sanapala says.

Gen 3 also features an all-new charging stand, which is designed for compatibility with all frame shapes and keeps lenses upright, protecting them from scratches while wirelessly charging.

Making smart eyewear look like eyewear

Making glasses that are suitable for everyday wear has always been a priority. “One of our goals has always been to develop technology that appears when you need it and disappears when you don’t,” says Wang.

Previous models of Echo Frames have come in a single, one-size-fits-all style.

A person is seen wearing Echo Frames sunglasses outside. The person carries a notebook and is looking down at it, and there are some buildings and blue sky in the background.
The Echo Frames team consulted with both internal and external eyewear designers to review common and popular styles of frames, and to survey potential customers about their preferences.

“That was a very intentional move,” Wang explains. “We wanted to start simply and learn from customer feedback.”

Gen 2’s flexible spring hinge and adjustable temple tips ensured that the single size fit many different faces. In fact, Wang says, while the goal was to fit around half of all potential users, they’ve found that 85 percent of the adult population can comfortably wear the Gen 2 design.

But with Gen 3, Wang says, the team needed to go beyond designing glasses that looked typical. Customers wanted glasses that looked stylish, too.

The team consulted with both internal and external eyewear designers to review common and popular styles of frames, as well as “edgier” designs, and to survey potential customers about their preferences. After testing options with beta customers, they settled on a variety of styles in various colors that cover a range of aesthetics. They also switched to an acetate material to match the feel of high-end eyewear.

Related content
How a team of designers, scientists, developers, and engineers worked together to create a truly unique device in Echo Show 10.

While each style will still come in a single size, the range of designs will accommodate even more faces than Gen 2, as the collection spans narrow, medium, and wide fits. Each style features adjustable temple tips constructed out of silicone around a lightweight titanium core for better fit. And despite the boost in battery life, the temples of Gen 3 frames have actually been slimmed down. Wang notes that competitive products often place large batteries behind a user’s ears. But presenting Echo Frames users with something that bulky and uncomfortable was never on the table.

“We were working with really heavy constraints,” Wang says. “So we have been very deliberate in making design choices in the service of our customer. That’s challenged us to be innovative and really push the limits of what’s possible in the architecture of our designs.”

Related content

US, MA, N.reading
Amazon is seeking exceptional talent to help develop the next generation of advanced robotics systems that will transform automation at Amazon's scale. We're building revolutionary robotic systems that combine cutting-edge AI, sophisticated control systems, and advanced mechanical design to create adaptable automation solutions capable of working safely alongside humans in dynamic environments. This is a unique opportunity to shape the future of robotics and automation at an unprecedented scale, working with world-class teams pushing the boundaries of what's possible in robotic dexterous manipulation, locomotion, and human-robot interaction. This role presents an opportunity to shape the future of robotics through innovative applications of deep learning and large language models. At Amazonwe leverage advanced robotics, machine learning, and artificial intelligence to solve complex operational challenges at an unprecedented scale. Our fleet of robots operates across hundreds of facilities worldwide, working in sophisticated coordination to fulfill our mission of customer excellence. The ideal candidate will contribute to research that bridges the gap between theoretical advancement and practical implementation in robotics. You will be part of a team that's revolutionizing how robots learn, adapt, and interact with their environment. Join us in building the next generation of intelligent robotics systems that will transform the future of automation and human-robot collaboration. Key job responsibilities - Collaborate with simulation and robotics experts to translate physical modeling needs into robust, scalable, and maintainable simulation solutions. - Design and implement high-performance simulation modeling and tools for rigid and deformable body simulation. - Identify and optimize performance bottlenecks in simulation pipelines to support real-time and batch simulation workflows. - Help build validation and unit testing pipelines to ensure correctness and physical fidelity of simulation results. - Identify potential sources of sim-to-real gaps and propose modeling and numerical approximations to reduce them. - Stay current with the latest advances in numerical methods, parallel computing, and GPU architectures, and incorporate them into our tools.
US, MA, North Reading
robotics systems that will transform automation at Amazon's scale. We're building revolutionary robotic systems that combine cutting-edge AI, sophisticated control systems, and advanced mechanical design to create adaptable automation solutions capable of working safely alongside humans in dynamic environments. This is a unique opportunity to shape the future of robotics and automation at unprecedented scale, working with world-class teams pushing the boundaries of what's possible in robotic manipulation, locomotion, and human-robot interaction. This role presents an opportunity to shape the future of robotics through innovative applications of deep learning and large language models. At Amazon Industrial Robotics we leverage advanced robotics, machine learning, and artificial intelligence to solve complex operational challenges at unprecedented scale. Our fleet of robots operates across hundreds of facilities worldwide, working in sophisticated coordination to fulfill our mission of customer excellence. We are pioneering the development of robotics foundation models that: Enable unprecedented generalization across diverse tasks Enable unprecedented robustness and reliability, industry-ready Integrate multi-modal learning capabilities (visual, tactile, linguistic) Accelerate skill acquisition through demonstration learning Enhance robotic perception and environmental understanding Streamline development processes through reusable capabilities The ideal candidate will contribute to research that bridges the gap between theoretical advancement and practical implementation in robotics. You will be part of a team that's revolutionizing how robots learn, adapt, and interact with their environment. Join us in building the next generation of intelligent robotics systems that will transform the future of automation and human-robot collaboration. As an Applied Science Manager in the Foundation Model team, you will build and lead a team that develops and improves machine learning systems that help robots perceive, reason, and act in real-world environments. You will set the technical direction for leveraging state-of-the-art models (open source and internal research), evaluating them on representative tasks, and adapting/optimizing them to meet robustness, safety, and performance needs. You will drive the capability roadmap and the evaluation strategy that defines “what the robot brain can do,” and you will sponsor targeted innovation when gaps remain. You’ll collaborate closely with research, controls, hardware, and product teams, and ensure the team’s outputs can be further customized and deployed by downstream teams on specific robot embodiments.
IN, KA, Bengaluru
Are you passionate about giving customers the richest, most inspiring experience in their shopping journey? Do you like to dive deep to understand how customer-centric solutions drive measurable results? Do you enjoy working closely with the business and software engineers to design rigorous experiments, build the data infrastructure behind them, and translate results into decisions? You are in the right place! Come join our Prime & Marketing Analytics and Science (PRIMAS) team, where your work will directly impact millions of customers. The EU Marketing & Prime organization is looking for a Data Scientist to join the PRIMAS team. This role sits at the intersection of applied statistics and large-scale analytics — you'll design experiments and causal models, and also own the data pipelines, metrics, and reporting infrastructure that make those results usable across the business. The PRIMAS team provides a comprehensive understanding of customer segments, affinities, and lifetime value. We use data science tools and advanced statistical techniques to study customer purchase and engagement behaviors, and generate actionable insights on where, when, and how we deliver products and programs to customers. We help increase customer engagement, sales, and marketing efficiency, and our systems are built entirely in-house on automated large-scale analytics infrastructure. You will design, launch, and measure experiments across marketing channels (SEM/SEO, Affiliates, Display, Social, Mobile, Email, Onsite, etc.), engagement products, and customer segments. You will improve our understanding of customer behavior, run rigorous power and minimum detectable effect (MDE) analyses to size experiments correctly, and build the causal and conversion models that value and target our marketing — then build the pipelines and dashboards that keep those signals flowing reliably to stakeholders and downstream systems. You will work at the forefront of consumer analytics, tackling some of the hardest measurement problems in the industry alongside strong scientists, statisticians, and software engineers. Key job responsibilities 1. Design and implement scalable, statistically rigorous experiments (A/B, geo, holdout, quasi-experiments) to measure marketing incrementality across channels. 2. Perform power analysis and minimum detectable effect (MDE) calculations to determine experiment sample sizes, durations, and design trade-offs before launch. 3. Build causal and treatment-effect models that produce conversion and valuation signals consumed by downstream bidding and budgeting systems. 4. Building the ETL, metric definitions, and datasets that make results scalable, extensible, and repeatable rather than one-off analyses. 5. Develop measurement frameworks that quantify the true, platform-independent contribution of marketing over time, and build the dashboards and reporting that keep those metrics visible to the business. 6. Apply statistical, mathematical, and machine learning techniques to solve ambiguous business problems where the right approach isn't obvious. 7. Analyze experiment results for validity — inspecting distributions, checking for sample ratio mismatch, exploring covariate balance, and tracking down the source of anomalies. 8. Communicate experiment design, results, and trade-offs clearly to business and leadership audiences, including inputs into business reviews, and influence decisions and technical direction across teams. 9. Establish scalable, repeatable processes and best practices for experiment design, data modeling, and analysis.
IN, KA, Bengaluru
Are you passionate about giving customers the richest, most inspiring experience in their shopping journey? Do you like to dive deep to understand how customer-centric solutions drive measurable results? Do you enjoy working closely with the business and software engineers to design rigorous experiments, build the data infrastructure behind them, and translate results into decisions? You are in the right place! Come join our Prime & Marketing Analytics and Science (PRIMAS) team, where your work will directly impact millions of customers. The EU Marketing & Prime organization is looking for a Data Scientist to join the PRIMAS team. This role sits at the intersection of applied statistics and large-scale analytics — you'll design experiments and causal models, and also own the data pipelines, metrics, and reporting infrastructure that make those results usable across the business. The PRIMAS team provides a comprehensive understanding of customer segments, affinities, and lifetime value. We use data science tools and advanced statistical techniques to study customer purchase and engagement behaviors, and generate actionable insights on where, when, and how we deliver products and programs to customers. We help increase customer engagement, sales, and marketing efficiency, and our systems are built entirely in-house on automated large-scale analytics infrastructure. You will design, launch, and measure experiments across marketing channels (SEM/SEO, Affiliates, Display, Social, Mobile, Email, Onsite, etc.), engagement products, and customer segments. You will improve our understanding of customer behavior, run rigorous power and minimum detectable effect (MDE) analyses to size experiments correctly, and build the causal and conversion models that value and target our marketing — then build the pipelines and dashboards that keep those signals flowing reliably to stakeholders and downstream systems. You will work at the forefront of consumer analytics, tackling some of the hardest measurement problems in the industry alongside strong scientists, statisticians, and software engineers. Key job responsibilities 1. Design and implement scalable, statistically rigorous experiments (A/B, geo, holdout, quasi-experiments) to measure marketing incrementality across channels. 2. Perform power analysis and minimum detectable effect (MDE) calculations to determine experiment sample sizes, durations, and design trade-offs before launch. 3. Build causal and treatment-effect models that produce conversion and valuation signals consumed by downstream bidding and budgeting systems. 4. Building the ETL, metric definitions, and datasets that make results scalable, extensible, and repeatable rather than one-off analyses. 5. Develop measurement frameworks that quantify the true, platform-independent contribution of marketing over time, and build the dashboards and reporting that keep those metrics visible to the business. 6. Apply statistical, mathematical, and machine learning techniques to solve ambiguous business problems where the right approach isn't obvious. 7. Analyze experiment results for validity — inspecting distributions, checking for sample ratio mismatch, exploring covariate balance, and tracking down the source of anomalies. 8. Communicate experiment design, results, and trade-offs clearly to business and leadership audiences, including inputs into business reviews, and influence decisions and technical direction across teams. 9. Establish scalable, repeatable processes and best practices for experiment design, data modeling, and analysis.
IN, KA, Bengaluru
Are you passionate about giving customers the richest, most inspiring experience in their shopping journey? Do you like to dive deep to understand how customer-centric solutions drive measurable results? Do you enjoy working closely with the business and software engineers to design rigorous experiments, build the data infrastructure behind them, and translate results into decisions? You are in the right place! Come join our Prime & Marketing Analytics and Science (PRIMAS) team, where your work will directly impact millions of customers. The EU Marketing & Prime organization is looking for a Data Scientist to join the PRIMAS team. This role sits at the intersection of applied statistics and large-scale analytics — you'll design experiments and causal models, and also own the data pipelines, metrics, and reporting infrastructure that make those results usable across the business. The PRIMAS team provides a comprehensive understanding of customer segments, affinities, and lifetime value. We use data science tools and advanced statistical techniques to study customer purchase and engagement behaviors, and generate actionable insights on where, when, and how we deliver products and programs to customers. We help increase customer engagement, sales, and marketing efficiency, and our systems are built entirely in-house on automated large-scale analytics infrastructure. You will design, launch, and measure experiments across marketing channels (SEM/SEO, Affiliates, Display, Social, Mobile, Email, Onsite, etc.), engagement products, and customer segments. You will improve our understanding of customer behavior, run rigorous power and minimum detectable effect (MDE) analyses to size experiments correctly, and build the causal and conversion models that value and target our marketing — then build the pipelines and dashboards that keep those signals flowing reliably to stakeholders and downstream systems. You will work at the forefront of consumer analytics, tackling some of the hardest measurement problems in the industry alongside strong scientists, statisticians, and software engineers. Key job responsibilities 1. Design and implement scalable, statistically rigorous experiments (A/B, geo, holdout, quasi-experiments) to measure marketing incrementality across channels. 2. Perform power analysis and minimum detectable effect (MDE) calculations to determine experiment sample sizes, durations, and design trade-offs before launch. 3. Build causal and treatment-effect models that produce conversion and valuation signals consumed by downstream bidding and budgeting systems. 4. Building the ETL, metric definitions, and datasets that make results scalable, extensible, and repeatable rather than one-off analyses. 5. Develop measurement frameworks that quantify the true, platform-independent contribution of marketing over time, and build the dashboards and reporting that keep those metrics visible to the business. 6. Apply statistical, mathematical, and machine learning techniques to solve ambiguous business problems where the right approach isn't obvious. 7. Analyze experiment results for validity — inspecting distributions, checking for sample ratio mismatch, exploring covariate balance, and tracking down the source of anomalies. 8. Communicate experiment design, results, and trade-offs clearly to business and leadership audiences, including inputs into business reviews, and influence decisions and technical direction across teams. 9. Establish scalable, repeatable processes and best practices for experiment design, data modeling, and analysis.
US, WA, Seattle
Want to apply data science to one of Amazon's fastest-growing payment businesses, serving millions of sellers and buyers across 20+ global marketplaces? As a Data Scientist I on the cross-border payments science team, you will build and deploy production models that power real-time FX risk monitoring, generative AI seller chatbots, and multi-agent AI tools. You will work end-to-end—from problem framing through production deployment on AWS—delivering solutions that directly influence billion-dollar payment flows. This is a high-visibility, small-team environment where your work informs senior leadership decisions and creates measurable impact for customers worldwide. Key job responsibilities - Build, tune, and evaluate large language models and generative AI applications, including seller-facing chatbots and multi-agent AI tools for the cross-border payments business. - Develop and deploy real-time statistical and machine learning models for FX risk monitoring, validating your data, assumptions, and results throughout the process. - Gather and use large datasets from multiple sources across global marketplaces to design production-ready solutions that meet customer needs and team goals. - Write accurate, clear, and mathematically rigorous technical documents that communicate model performance and business impact to both technical and non-technical audiences. - Partner with engineering, business, and science teams to translate payment-domain problems into data science solutions and ensure smooth deployment on AWS. A day in the life You will spend your morning reviewing model outputs from production FX risk systems and tuning generative AI prototypes for seller chatbots. In the afternoon you might pair with an engineer to deploy a new model version on AWS, then present preliminary results to senior leadership. Throughout, you will seek feedback from senior scientists on your methodology and collaborate with cross-functional partners to refine problem framing. About the team We are the science team behind Amazon's cross-border payments business, supporting products that serve millions of sellers and buyers across 20+ global marketplaces. Our team is small and at an inflection point—we are expanding our generative AI capabilities, building multi-agent systems, and strengthening real-time risk models. You will join a group that values end-to-end ownership, from research to production, and whose work directly shapes decisions on large-scale payment flows.
US, TX, Austin
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
US, WA, Bellevue
The Amazon GDS-MOP (modeling, Optimization and Planning) Science team is seeking an exceptional Applied Scientist with strong operations research and optimization expertise to develop production solutions for one of the most complex systems in the world: Amazon's Fulfillment Network labor capacity planning. At MOP Science, we design, build, and deploy optimization, statistics, machine learning, and GenAI/LLM solutions that power Amazon Labor Planning systems (ALPS) running across Amazon Fulfillment Centers worldwide. We solve a wide range of challenges encountered throughout the network, including labor planning and staffing, pick scheduling, stow guidance, and capacity risk management. We are tasked with developing innovative, scalable, and reliable science-driven production solutions that exceed the published state of the art, enabling systems to run frequently (ranging from every few minutes to every few hours per use case) and continuously across our large-scale network. Key job responsibilities As an Applied Scientist, you will collaborate with other scientists, software engineers, product managers, and operations leaders to develop optimization-driven solutions using a variety of tools and observe direct impact on process efficiency and associate experience in the fulfillment network. Key responsibilities include: • Develop understanding and domain knowledge of operational processes, system architecture and functions, and business requirements • Deep dive into data and code to identify opportunities for continuous improvement and/or disruptive new approaches • Develop scalable mathematical models for production systems to derive optimal or near-optimal solutions for existing and new challenges • Create prototypes and simulations for agile experimentation of devised solutions • Advocate for technical solutions with business stakeholders, engineering teams, and senior leadership • Partner with engineers to integrate prototypes into production systems • Design experiments to test new or incremental solutions launched in production and build metrics to track performance About the team Amazon offers a full range of benefits that support you and eligible family members, including domestic partners and their children. Benefits can vary by location, the number of regularly scheduled hours you work, length of employment, and job status such as seasonal or temporary employment. The benefits that generally apply to regular, full-time employees include: • Medical, Dental, and Vision Coverage • Maternity and Parental Leave Options • Paid Time Off (PTO) • 401(k) Plan
US, CA, Sunnyvale
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
US, WA, Redmond
We are searching for a talented candidate with experience in orbital mechanics, orbit determination, launch vehicle trajectories, and launch vehicle mission planning. In this position, a successful candidate would serve as a Research Scientist in support of Amazon Leo’s constellation with particular focus on Launch Vehicle support. Strong analysis skills are required to develop engineering studies of complex large-scale dynamical systems. This position requires demonstrated expertise in computational analysis automation and tool development. 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 Working with the Leo GNC team, you will: • Perform spacecraft maneuver or navigation analysis in support of multi-disciplinary trades within the Amazon Leo team. • Contribute to prototype software development of flight algorithms. • Test and assess navigation software for integration into flight systems. • Assess and trouble-shoot the performance of Leo on-board GNSS hardware and software systems. • Work closely with GNC engineers to manage on-orbit performance and develop flight dynamics operations processes. • Manage engineering trades as needed for various launch vehicle mission designs. • Support Leo’s Launch Vehicle Mission Management team with technical expertise in Launch Vehicle trajectory requirements specification • Evaluate Launch Vehicle performance and compliance with mission requirements • Develop tools to support Mission Management planning for over 80 launches! • Work collaboratively with launch vehicle system technical teams About the team The Flight Dynamics team is responsible for the guidance, navigation, control and safety of the spaceflight of the Amazon Leo constellation. This team provides solutions to spaceflight challenges in constellation design, orbit selection, launch vehicle insertion requirements, navigation, trajectory design, space situational awareness, and space traffic coordination.