Events & Conferences
How Amazon scientists are driving success for Alexa in the car

“Alexa, where’s the nearest coffee shop?”
In vehicles with Alexa, drivers can pose questions like that — while keeping their eyes on the road and hands on the wheel. Amazon’s cloud-based voice assistant technology works with the car’s navigation system to figure out where the nearest coffee shop is and guide drivers to it.
Interactions like this hinge on the ability of the built-in hardware and software to accurately detect when Alexa is invoked.
“We want to take the Alexa experience that customers are familiar and comfortable with in the home and extend that experience to cars,” said Jinghao Liu, a principal hardware engineer with Amazon Smart Vehicles at the team’s vehicle lab in San Jose, California.
The challenge is that cars are a different acoustic and operating environment than a home. They are noisier, for starters. What’s more, the bulk of customer needs in a car are location-specific and tied to real-time road, traffic, and weather conditions. And, as most drivers know, cellular service in a car is never 100% reliable, especially outside of cities and towns.
There’s another complication: the hardware. Amazon scientists often work with car manufacturers on the design and implementation of their systems, but during validation in the 6,000-square-foot vehicle lab, there is one element they cannot control for.
“We are giving our partners access to Alexa services and functionalities, but we don’t make the device,” explained Parimal Tathavadekar, a senior manager on the Amazon Smart Vehicles team. “We make sure that the integration of our technology with their device is correct and happily working.”
Tackling road noise
Perhaps the biggest difference between speaking to Alexa in a car versus at home is the increase in noise, Liu noted. Think of the rumble from tires on the road and turbulence from the wind, as well as whatever radio station, podcast, or music is playing through the stereo.
To invoke Alexa, the car’s microphones need to pick out the speaker’s voice from this din and send the voice signal to the automatic speech recognition model running in the cloud.
In design meetings, Amazon scientists collaborate with car manufacturers on techniques that allow the car’s built-in system to focus on the person speaking to Alexa and cancel out acoustic echo and background noise.
For example, Amazon scientists recommend car manufacturers use a signal-processing technique called beamforming that steers microphone arrays toward a voice signal coming from a particular source, such as the driver, while suppressing audio interference from all other directions.
“You are laser focusing on just that one angle,” Liu explained, adding that beamforming is used in the Amazon Echo family of devices.
Another recommended technique called acoustic echo cancellation blocks out the noise from the car’s speakers so that music played by the speaker does not interfere with the customer requests. Manufacturers can also deploy various noise reduction algorithms, which learn to separate a speaker’s voice from other background noise.
Inside the vehicle lab, Amazon scientists use software and acoustic tools to simulate road noise, music, and other acoustic elements as they test and validate Alexa integration with the car manufacturer’s built-in systems.
To achieve certification, the built-in system must stream a voice signal to the automatic-speech-recognition model that’s of a similar quality to the voice signals streamed from any of the other hundreds of millions of Alexa-connected devices around the world.
“This model is serving millions of devices at home and on the go, so we make sure that the received audio signal is reasonable, and our cloud takes care of the rest,” Liu said.
Personal mobility
Customer interactions with Alexa in cars are often location-specific, ranging from queries about points of interest along the way to driving directions that avoid traffic jams.
Amazon does not have its own navigation system or a proprietary database of points of interest, noted Tathavadekar.
Rather, Alexa AI is designed to resolve a speaker’s query and hand off such intents to the car’s navigation engine for geolocation information. Alexa uses this data to interact with third-party databases for information about points of interest.
The use of Alexa in conjunction with a car’s native navigation system enables drivers to accomplish tasks via voice instead of “having to type or touch through different navigation screens, which can be a distraction,” Tathavadekar said.
The contextual awareness also allows Alexa to deliver experiences such as tuning in to local radio stations on a cross-country road trip.
Inside Amazon’s vehicle lab, engineers simulate vehicle locations, Liu explained, validating that Alexa is integrated with the car’s navigation system and able to provide relevant information and experiences to customers on the go.
Staying connected
Quality of connection is another factor in how quickly a device can connect to Alexa and get a response.
“When you’re driving, you’re going to go through a patch with poor reception,” Liu noted. That’s why he and his colleagues also test how built-in systems handle scenarios with poor, or no, cellular signal.
In today’s world, customers are used to getting a quick response from Alexa at home, Liu noted. If the response takes longer on the road, the customer may think the device is defective. That’s why Amazon makes sure that Alexa stays connected to customers, even when the cellular signal is lost.
For example, Alexa’s local voice control of native features in the car such as adjusting the air conditioning, tuning the radio, opening the sunroof, and turning on the reading light do not require a connection to the cloud.
In the vehicle lab, the team measures the built-in hardware device’s latency with various levels of connectivity. If the cellular signal strength is strong, but Alexa is slow to respond, that may be an indication that the hardware needs to be re-engineered.
As connectivity becomes ubiquitous, Liu added, the latency expectations will continue to tighten. Amazon scientists also expect that experiences within and outside of cars will become more engaging and complex going forward.
“Every year, we’re adding new features and equipment to our test arsenal,” Liu said.
Making it personal
The Alexa experience in a car should also be as personal as the interactions that customers have with Alexa in their homes, noted Tathavadekar. That’s possible because cars with built-in Alexa are just another endpoint for individual Alexa accounts, which are stored in the cloud.
“We are able to render experiences on all the endpoints where that particular person might be operating,” he said. “You might create a playlist at home and play it in your car.”
In a similar fashion, a driver in their car can ask Alexa to add an item to a shopping list that they started in the kitchen or ask Alexa to lock the front door of a connected smart home.
Providing that personal and familiar experience with Alexa in cars requires a system that seamlessly extends the Alexa experience from an Echo at home to a vehicle that might be traveling down a congested highway with patchy connectivity.
“The big chunk of our work is to collaborate with the major car manufacturers to create unique experiences that delight car owners,” said Liu. “They are our partners who help deliver the product to customers. They’re making the car, and we add this piece of personalized voice assistant into the digital experience.”
Events & Conferences
A New Ranking Framework for Better Notification Quality on Instagram

- We’re sharing how Meta is applying machine learning (ML) and diversity algorithms to improve notification quality and user experience.
- We’ve introduced a diversity-aware notification ranking framework to reduce uniformity and deliver a more varied and engaging mix of notifications.
- This new framework reduces the volume of notifications and drives higher engagement rates through more diverse outreach.
Notifications are one of the most powerful tools for bringing people back to Instagram and enhancing engagement. Whether it’s a friend liking your photo, another close friend posting a story, or a suggestion for a reel you might enjoy, notifications help surface moments that matter in real time.
Instagram leverages machine learning (ML) models to decide who should get a notification, when to send it, and what content to include. These models are trained to optimize for user positive engagement such as click-through-rate (CTR) – the probability of a user clicking a notification – as well as other metrics like time spent.
However, while engagement-optimized models are effective at driving interactions, there’s a risk that they might overprioritize the product types and authors someone has previously engaged with. This can lead to overexposure to the same creators or the same product types while overlooking other valuable and diverse experiences.
This means people could miss out on content that would give them a more balanced, satisfying, and enriched experience. Over time, this can make notifications feel spammy and increase the likelihood that people will disable them altogether.
The real challenge lies in finding the right balance: How can we introduce meaningful diversity into the notification experience without sacrificing the personalization and relevance people on Instagram have come to expect?
To tackle this, we’ve introduced a diversity-aware notification ranking framework that helps deliver more diverse, better curated, and less repetitive notifications. This framework has significantly reduced daily notification volume while improving CTR. It also introduces several benefits:
- The extensibility of incorporating customized soft penalty (demotion) logic for each dimension, enabling more adaptive and sophisticated diversity strategies.
- The flexibility of tuning demotion strength across dimensions like content, author, and product type via adjustable weights.
- The integration of balancing personalization and diversity, ensuring notifications remain both relevant and varied.
The Risks of Notifications without Diversity
The issue of overexposure in notifications often shows up in two major ways:
Overexposure to the same author: People might receive notifications that are mostly about the same friend. For example, if someone often interacts with content from a particular friend, the system may continue surfacing notifications from that person alone – ignoring other friends they also engage with. This can feel repetitive and one-dimensional, reducing the overall value of notifications.
Overexposure to the same product surface: People might mostly receive notifications from the same product surface such as Stories, even when Feed or Reels could provide value. For example, someone may be interested in both reel and story notifications but has recently interacted more often with stories. Because the system heavily prioritizes past engagement, it sends only story notifications, overlooking the person’s broader interests.
Introducing Instagram’s Diversity-Aware Notification Ranking Framework
Instagram’s diversity-aware notification ranking framework is designed to enhance the notification experience by balancing the predicted potential for user engagement with the need for content diversity. This framework introduces a diversity layer on top of the existing engagement ML models, applying multiplicative penalties to the candidate scores generated by these models, as figure1, below, shows.
The diversity layer evaluates each notification candidate’s similarity to recently sent notifications across multiple dimensions such as content, author, notification type, and product surface. It then applies carefully calibrated penalties—expressed as multiplicative demotion factors—to downrank candidates that are too similar or repetitive. The adjusted scores are used to re-rank the candidates, enabling the system to select notifications that maintain high engagement potential while introducing meaningful diversity. In the end, the quality bar selects the top-ranked candidate that passes both the ranking and diversity criteria.
Mathematical Formulation
Within the diversity layer, we apply a multiplicative demotion factor to the base relevance score of each candidate. Given a notification candidate 𝑐, we compute its final score as the product of its base ranking score and a diversity demotion multiplier:
where R(c) represents the candidate’s base relevance score, and D(c) ∈ [0,1] is a penalty factor that reduces the score based on similarity to recently sent notifications. We define a set of semantic dimensions (e.g., author, product type) along which we want to promote diversity. For each dimension i, we compute a similarity signal pi(c) between candidate c and the set of historical notifications H, using a maximal marginal relevance (MMR) approach:
where simi(·,·) is a predefined similarity function for dimension i. In our baseline implementation, pi(c) is binary: it equals 1 if the similarity exceeds a threshold 𝜏i and 0 otherwise.
The final demotion multiplier is defined as:
where each wi ∈ [0,1] controls the strength of demotion for its respective dimension. This formulation ensures that candidates similar to previously delivered notifications along one or more dimensions are proportionally down-weighted, reducing redundancy and promoting content variation. The use of a multiplicative penalty allows for flexible control across multiple dimensions, while still preserving high-relevance candidates.
The Future of Diversity-Aware Ranking
As we continue evolving our notification diversity-aware ranking system, a next step is to introduce more adaptive, dynamic demotion strategies. Instead of relying on static rules, we plan to make demotion strength responsive to notification volume and delivery timing. For example, as a user receives more notifications—especially of similar type or in rapid succession—the system progressively applies stronger penalties to new notification candidates, effectively mitigating overwhelming experiences caused by high notification volume or tightly spaced deliveries.
Longer term, we see an opportunity to bring large language models (LLMs) into the diversity pipeline. LLMs can help us go beyond surface-level rules by understanding semantic similarity between messages and rephrasing content in more varied, user-friendly ways. This would allow us to personalize notification experiences with richer language and improved relevance while maintaining diversity across topics, tone, and timing.
Events & Conferences
Simplifying book discovery with ML-powered visual autocomplete suggestions

Every day, millions of customers search for books in various formats (audiobooks, e-books, and physical books) across Amazon and Audible. Traditional keyword autocomplete suggestions, while helpful, usually require several steps before customers find their desired content. Audible took on the challenge of making book discovery more intuitive and personalized while reducing the number of steps to purchase.
We developed an instant visual autocomplete system that enhances the search experience across Amazon and Audible. As the user begins typing a query, our solution provides visual previews with book covers, enabling direct navigation to relevant landing pages instead of the search result page. It also delivers real-time personalized format recommendations and incorporates multiple searchable entities, such as book pages, author pages, and series pages.
Our system needed to understand user intent from just a few keystrokes and determine the most relevant books to display, all while maintaining low latency for millions of queries. Using historical search data, we match keystrokes to products, transforming partial inputs into meaningful search suggestions. To ensure quality, we implemented confidence-based filtering mechanisms, which are particularly important for distinguishing between general queries like “mystery” and specific title searches. To reflect customers’ most recent interests, the system applies time-decay functions to long historical user interaction data.
To meet the unique requirements of each use case, we developed two distinct technical approaches. On Audible, we deployed a deep pairwise-learning-to-rank (DeepPLTR) model. The DeepPLTR model considers pairs of books and learns to assign a higher score to the one that better matches the customer query.
The DeepPLTR model’s architecture consists of three specialized towers. The left tower factors in contextual features and recent search patterns using a long-short-term-memory model, which processes data sequentially and considers its prior decisions when issuing a new term in the sequence. The middle tower handles keyword and item engagement history. The right tower factors in customer taste preferences and product descriptions to enable personalization. The model learns from paired examples, but at runtime, it relies on books’ absolute scores to assemble a ranked list.
For Amazon, we implemented a two-stage modeling approach involving a probabilistic information-retrieval model to determine the book title that best matches each keyword and a second model that personalizes the book format (audiobooks, e-books, and physical books). This dual-strategy approach maintains low latency while still enabling personalization.
In practice, a customer who types “dungeon craw” in the search bar now sees a visual recommendation for the book Dungeon Crawler Carl, complete with book cover, reducing friction by bypassing a search results page and sending the customer directly to the product detail page. On Audible, the system also personalizes autocomplete results and enriches the discovery experience with relevant connections. These include links to the author’s complete works (Matt Dinniman’s author page) and, for titles that belong to a series, links to the full collection (such as the Dungeon Crawler Carl series).
On Amazon, when the customer clicks on the title, the model personalizes the right book-format (audiobooks, e-books, physical books) recommendation and directs the customer to the right product detail page.
In both cases, after the customer has entered a certain number of keystrokes, the system employs a model to detect customer intent (e.g., book title intent for Amazon or author intent for Audible) and determine which visual widget should be displayed.
Audible and Amazon books’ visual autocomplete provides customers with more relevant content more rapidly than traditional autocomplete, and its direct navigation reduces the number of steps to find and access desired books — all while handling millions of queries at low latency.
This technology is not just about making book discovery easier; it is laying the foundation for future improvements in search personalization and visual discovery across Amazon’s ecosystem.
Acknowledgements: Jiun Kim, Sumit Khetan, Armen Stepanyan, Jack Xuan, Nathan Brothers, Eddie Chen, Vincent Lee, Soumy Ladha, Justine Luo, Yuchen Zeng, David Torres, Gali Deutsch, Chaitra Ramdas, Christopher Gomez, Sharmila Tamby, Melissa Ma, Cheng Luo, Jeffrey Jiang, Pavel Fedorov, Ronald Denaux, Aishwarya Vasanth, Azad Bajaj, Mary Heer, Adam Lowe, Jenny Wang, Cameron Cramer, Emmanuel Ankrah, Lydia Diaz, Suzette Islam, Fei Gu, Phil Weaver, Huan Xue, Kimmy Dai, Evangeline Yang, Chao Zhu, Anvy Tran, Jessica Wu, Xiaoxiong Huang, Jiushan Yang
Events & Conferences
Revolutionizing warehouse automation with scientific simulation

Modern warehouses rely on complex networks of sensors to enable safe and efficient operations. These sensors must detect everything from packages and containers to robots and vehicles, often in changing environments with varying lighting conditions. More important for Amazon, we need to be able to detect barcodes in an efficient way.
The Amazon Robotics ID (ARID) team focuses on solving this problem. When we first started working on it, we faced a significant bottleneck: optimizing sensor placement required weeks or months of physical prototyping and real-world testing, severely limiting our ability to explore innovative solutions.
To transform this process, we developed Sensor Workbench (SWB), a sensor simulation platform built on NVIDIA’s Isaac Sim that combines parallel processing, physics-based sensor modeling, and high-fidelity 3-D environments. By providing virtual testing environments that mirror real-world conditions with unprecedented accuracy, SWB allows our teams to explore hundreds of configurations in the same amount of time it previously took to test just a few physical setups.
Camera and target selection/positioning
Sensor Workbench users can select different cameras and targets and position them in 3-D space to receive real-time feedback on barcode decodability.
Three key innovations enabled SWB: a specialized parallel-computing architecture that performs simulation tasks across the GPU; a custom CAD-to-OpenUSD (Universal Scene Description) pipeline; and the use of OpenUSD as the ground truth throughout the simulation process.
Parallel-computing architecture
Our parallel-processing pipeline leverages NVIDIA’s Warp library with custom computation kernels to maximize GPU utilization. By maintaining 3-D objects persistently in GPU memory and updating transforms only when objects move, we eliminate redundant data transfers. We also perform computations only when needed — when, for instance, a sensor parameter changes, or something moves. By these means, we achieve real-time performance.
Visualization methods
Sensor Workbench users can pick sphere- or plane-based visualizations, to see how the positions and rotations of individual barcodes affect performance.
This architecture allows us to perform complex calculations for multiple sensors simultaneously, enabling instant feedback in the form of immersive 3-D visuals. Those visuals represent metrics that barcode-detection machine-learning models need to work, as teams adjust sensor positions and parameters in the environment.
CAD to USD
Our second innovation involved developing a custom CAD-to-OpenUSD pipeline that automatically converts detailed warehouse models into optimized 3-D assets. Our CAD-to-USD conversion pipeline replicates the structure and content of models created in the modeling program SolidWorks with a 1:1 mapping. We start by extracting essential data — including world transforms, mesh geometry, material properties, and joint information — from the CAD file. The full assembly-and-part hierarchy is preserved so that the resulting USD stage mirrors the CAD tree structure exactly.
To ensure modularity and maintainability, we organize the data into separate USD layers covering mesh, materials, joints, and transforms. This layered approach ensures that the converted USD file faithfully retains the asset structure, geometry, and visual fidelity of the original CAD model, enabling accurate and scalable integration for real-time visualization, simulation, and collaboration.
OpenUSD as ground truth
The third important factor was our novel approach to using OpenUSD as the ground truth throughout the entire simulation process. We developed custom schemas that extend beyond basic 3-D-asset information to include enriched environment descriptions and simulation parameters. Our system continuously records all scene activities — from sensor positions and orientations to object movements and parameter changes — directly into the USD stage in real time. We even maintain user interface elements and their states within USD, enabling us to restore not just the simulation configuration but the complete user interface state as well.
This architecture ensures that when USD initial configurations change, the simulation automatically adapts without requiring modifications to the core software. By maintaining this live synchronization between the simulation state and the USD representation, we create a reliable source of truth that captures the complete state of the simulation environment, allowing users to save and re-create simulation configurations exactly as needed. The interfaces simply reflect the state of the world, creating a flexible and maintainable system that can evolve with our needs.
Application
With SWB, our teams can now rapidly evaluate sensor mounting positions and verify overall concepts in a fraction of the time previously required. More importantly, SWB has become a powerful platform for cross-functional collaboration, allowing engineers, scientists, and operational teams to work together in real time, visualizing and adjusting sensor configurations while immediately seeing the impact of their changes and sharing their results with each other.
New perspectives
In projection mode, an explicit target is not needed. Instead, Sensor Workbench uses the whole environment as a target, projecting rays from the camera to identify locations for barcode placement. Users can also switch between a comprehensive three-quarters view and the perspectives of individual cameras.
Due to the initial success in simulating barcode-reading scenarios, we have expanded SWB’s capabilities to incorporate high-fidelity lighting simulations. This allows teams to iterate on new baffle and light designs, further optimizing the conditions for reliable barcode detection, while ensuring that lighting conditions are safe for human eyes, too. Teams can now explore various lighting conditions, target positions, and sensor configurations simultaneously, gleaning insights that would take months to accumulate through traditional testing methods.
Looking ahead, we are working on several exciting enhancements to the system. Our current focus is on integrating more-advanced sensor simulations that combine analytical models with real-world measurement feedback from the ARID team, further increasing the system’s accuracy and practical utility. We are also exploring the use of AI to suggest optimal sensor placements for new station designs, which could potentially identify novel configurations that users of the tool might not consider.
Additionally, we are looking to expand the system to serve as a comprehensive synthetic-data generation platform. This will go beyond just simulating barcode-detection scenarios, providing a full digital environment for testing sensors and algorithms. This capability will let teams validate and train their systems using diverse, automatically generated datasets that capture the full range of conditions they might encounter in real-world operations.
By combining advanced scientific computing with practical industrial applications, SWB represents a significant step forward in warehouse automation development. The platform demonstrates how sophisticated simulation tools can dramatically accelerate innovation in complex industrial systems. As we continue to enhance the system with new capabilities, we are excited about its potential to further transform and set new standards for warehouse automation.
-
Business6 days ago
The Guardian view on Trump and the Fed: independence is no substitute for accountability | Editorial
-
Tools & Platforms3 weeks ago
Building Trust in Military AI Starts with Opening the Black Box – War on the Rocks
-
Ethics & Policy1 month ago
SDAIA Supports Saudi Arabia’s Leadership in Shaping Global AI Ethics, Policy, and Research – وكالة الأنباء السعودية
-
Events & Conferences4 months ago
Journey to 1000 models: Scaling Instagram’s recommendation system
-
Jobs & Careers2 months ago
Mumbai-based Perplexity Alternative Has 60k+ Users Without Funding
-
Education2 months ago
VEX Robotics launches AI-powered classroom robotics system
-
Funding & Business2 months ago
Kayak and Expedia race to build AI travel agents that turn social posts into itineraries
-
Podcasts & Talks2 months ago
Happy 4th of July! 🎆 Made with Veo 3 in Gemini
-
Podcasts & Talks2 months ago
OpenAI 🤝 @teamganassi
-
Education2 months ago
Macron says UK and France have duty to tackle illegal migration ‘with humanity, solidarity and firmness’ – UK politics live | Politics