Formally verified AES-XTS: The first AES algorithm to join s2n-bignum

Simplifying and clarifying the assembly code for core operations enabled automated optimization and verification.

Overview by Amazon Nova
  • AWS formally verified an optimized Arm64 assembly implementation of AES-XTS decryption, ensuring mathematical correctness and security.
  • The verification process involved simplifying the assembly code, using the HOL Light interactive theorem prover, and integrating formal verification into the continuous-integration workflow.
  • The formally verified AES-XTS implementation provides mathematical guarantees of correctness, enabling performance improvements and maintaining security for AWS's storage encryption algorithm.
Was this answer helpful?

Cryptographic encryption algorithms are mathematical procedures that transform readable data into ciphertext that looks like a stream of random bits. The ciphertext can be decrypted only with the corresponding decryption algorithm and the correct key.

For data at rest — information stored on disks or in databases — algorithms like AES-XTS encrypt each block before it’s written to storage, protecting against physical theft or unauthorized access to storage systems. For data in transit — information traveling across networks — protocols like TLS combine multiple algorithms: asymmetric encryption algorithms (RSA or elliptic curves) establish secure connections, while fast symmetric encryption algorithms (like AES-GCM) protect the actual data stream and verify that it hasn't been tampered with. At Amazon Web Services (AWS), we use AES-XTS to protect customer data in services like EBS, Nitro cards, and DynamoDB, while TLS with AES-GCM secures all network communication between services and to customers.

We took on the challenge of formally verifying an optimized Arm64 assembly implementation of AES-XTS decryption, where “formal verification” is the process of proving mathematically that an engineered system meets a particular specification.

Our work follows the IEEE Standard 1619 for cryptographic protection of block-oriented storage devices and focuses on the AES-256-XTS variant of AES-XTS. The “256” specifies the size of the encryption key.

Unlike algorithms that process fixed-size blocks, AES-XTS handles variable-length data from 16 bytes up to 16 megabytes, with special logic for incomplete blocks. The assembly code verified was a 5x-unrolled version, meaning that its loops were executed in parallel across five registers (each containing an input block), and it had been optimized for modern CPU pipelines. It was complex enough that manual review couldn't guarantee correctness yet critical enough that errors could compromise customer data security.

As part of Amazon Web Services’ s2n-bignum library of formally verified big-number operations, we contributed an improved Arm64 assembly implementation of AES-XTS encryption and decryption, as well as specification and formal verification using the HOL Light interactive theorem prover, which was developed by a member of our team (John Harrison).

This was an experiment in the proof-driven development of a large function with multiple paths based on the input length. It resulted in the largest proof so far in the s2n-bignum library. For the typical input size of 512 bytes, the performance of the algorithm either stayed close to that of original code or improved slightly on highly optimized Arm cores. By adding this algorithm and its proof to the s2n-bignum library, we pave the way for more AES-based algorithms to be added.

Description of the algorithm

AES is a block cipher that implements a keyed permutation. This means that it processes the plaintext files in blocks (in this case, blocks of 128 bits), and for any given key, it defines a bijective (one-to-one and invertible) function mapping each plaintext block to a unique ciphertext block. This mathematical property ensures that decryption can uniquely recover the original plaintext.

AES-XTS is the mode specifically designed for storage encryption. It uses AES as its underlying building block but adds position-dependent tweaks and ciphertext stealing — a method for handling partial blocks — to address the unique requirements of disk encryption, where you need random access to any sector and must preserve the exact data size.

AES-XTS encrypts storage data using a two-key approach where each 128-bit block and its position-dependent tweak are subjected to an exclusive-OR operation (XOR), a binary operation that outputs a one only if the input values differ. The result of the operation is encrypted with AES, then XORed with the tweak again, ensuring that identical data at different disk locations produces different ciphertext. The tweak is generated by encrypting the sector number with a second key, then multiplying by powers of α in a Galois field, creating unique values for each block position.

When the final block isn't a full 128 bits, ciphertext stealing kicks in. Ciphertext stealing borrows bytes from the previous block, allowing encryption of data of any length without padding or wasted space. This lets you read or write any sector independently — critical for disk encryption — while basing each block's encryption on its position. That is a desired feature since the security model of disk encryption allows the adversary to access sectors other than those in question, modify them, and request their decryption. It also ensures that the size of the ciphertext is exactly the same as that of the plaintext, so it fits in its place.

The basic XEX block encryption structure with the XOR-Encrypt-XOR flow v3.jpg
AES-XTS encryption employing an XOR-encrypt-XOR (XEX) structure.
Ciphertext-stealing-encryption.png
Ciphertext stealing handles partial blocks during encryption by splitting the second-to-last block's output.
Decryption.png
The symmetric decryption process.
Reverse ciphertext stealing for decryption.png
Reverse ciphertext stealing for decryption.

Control flow of the assembly implementation

We started from an existing implementation of AES-XTS in Amazon’s AWS-LC cryptographic library. AES-XTS loops through the plaintext in 128-bit blocks, and encryption of each block requires 15 steps, each with its own “round key” derived from the encryption key. The existing implementation is 5x unrolled, meaning it processes blocks in parallel, five at a time. If the final block is less than 128 bits in length, there’s a risk of “buffer overread”, or reading beyond the limits of the input buffer.

To avoid overread, the existing implementation does complex manipulation over the pointer to the current location in the input buffer. This requires a sophisticated control flow that can be hard to follow: the loop counter is incremented and decremented multiple times before and during the loop, and the loop has two additional exit points other than the final loop-back branch.

One exit point is for the case when four blocks remain during the final iteration of the loop; the other exit point is for the case of one to three blocks remaining. The flow of the loop interleaves the load/store instructions with the AES and XOR instructions in an effort to maximize pipeline usage. After the loop exit, the processing of the remaining blocks is intertwined for the lengths of four down to one; if there’s a partial block at the end, the algorithm performs the ciphertext-stealing procedure. Additionally, only seven of the 15 rounds’ keys were kept in registers; the other eight were repeatedly loaded from memory in the loop and outside it.

We first investigated whether we could improve the performance of the main loop by letting SLOTHY, a superoptimizer for Arm code, rearrange the instructions to maximize pipeline usage. SLOTHY contains simplified models of various Arm microarchitectures. It uses a constraint solver to provide optimal instruction schedule, register renaming, and periodic loop interleaving.

However, for SLOTHY to identify and optimize a loop, the loop has to exhibit typical loop behavior, decreasing the counter at the end of each iteration and then jumping back to the beginning. SLOTHY also cannot handle the nested loop created by loading the eight-round keys.

This gave us a reason to start “cleaning up” the loop. First, we freed up registers to permanently hold all round keys; this was possible as the optimized order of instructions required fewer temporary registers than the original code. Second, we removed the multiple exit points and the manipulation of the loop counter to handle the remaining blocks. The value of the counter always indicates the number of five-block chunks remaining in the buffer, conforming to SLOTHY’s requirement; the loop ends before the handling of the remaining blocks.

Once the loop ends, we have a separate processing branch to handle each possible number of remaining blocks, from one to four; all four branches end in ciphertext stealing. This can be seen in the control flow graphs of the encrypt and decrypt algorithms (see below). Throughout the code, we maintained the constant-time design mindset; that is, branching and special processing are based not on secret data but only on public values, the input byte lengths.

AES-256-XTS encryption control flow graph v2.jpg
AES-256-XTS encryption control flow graph.
AES-256-XTS decryption control flow graph v2.jpg
AES-256-XTS decryption control flow graph.

Performance

With our modifications to the code, we were able to use SLOTHY to optimize the encrypt algorithm. This resulted in slight performance gains on the AWS Graviton family of Arm processors, although the gains were smaller on the more advanced chips, which have an optimized out-of-order pipeline. Compared to the original implementation, keeping round keys in registers throughout the algorithm’s execution, to save on loading them from memory, allowed us to offset the effects of not interleaving the AES instructions with other ones.

Having a cleaner flow of instructions in the loop and modular exit processing allowed us to experiment with various unrolling factors for the loop iterations. We experimented with 3x, 4x, and 6x factors and concluded that 5x is still the best choice across various microarchitectures.

Ensuring correctness through formal verification

To deploy optimized cryptographic code in production, we need mathematical certainty that it works correctly. While random testing quickly checks simple cases, we rely on formal verification to deliver the highest level of assurance for our AES-XTS implementation.

Why HOL Light for AES-XTS?

To prove that our implementation matches the IEEE 1619 specification, we use HOL Light, an interactive theorem prover developed by our colleague John Harrison. HOL Light is a particularly simple implementation of the "correct by construction" approach to software development, in which code is verified as it’s written. HOL Light’s trusted kernel is just a few hundred lines of code, which implements basic logical inference rules. This means that even if there's a bug in our proof tactics or automation, it cannot cause HOL Light to accept an incorrect proof. At worst, a bug prevents us from completing a proof, but it cannot make a false statement provable.

We chose HOL Light for several reasons specific to AES-XTS verification:

  • Assembly-level verification: We write our implementations directly in assembly rather than relying on compiled code. While more challenging, this makes our proofs independent of any compiler. HOL Light reasons directly about machine code bytes using CPU instruction specifications, providing assurance at the lowest level of the software stack.
  • Existing cryptographic infrastructure: S2n-bignum already provides extensive support for cryptographic verification, including symbolic simulation that strips away execution artifacts and leaves purely mathematical problems, specialized tactics for word operations, and byte list handling. We add proven lemmas about AES operations that we can reuse for the proofs of other AES modes.
  • Complex control flow handling: Unlike fully automated provers that might fail on complex proofs without enough explanation, HOL Light's interactive approach lets us guide proofs through the intricate invariants required for our 5x-unrolled loops, processing arbitrarily long blocks of data and performing the complex memory reasoning required by variable-length inputs and partial blocks.

The s2n-bignum framework

Using s2n-bignum to implement AES-XTS serves two purposes: it's both a framework for formally verifying assembly code in x86-64 and Arm architectures and a collection of fast, verified assembly functions for cryptography. The library already contains verified implementations of numerous cryptographic algorithms, especially those pertaining to big-number mathematical operations (hence the name), which are the foundation of public-key cryptographic primitives. For details on how HOL Light was used to prove public-key algorithms as part of s2n-bignum, please refer to the previous Amazon Science blog posts “ Formal verification makes RSA faster — and faster to deploy” and “Better-performing ‘25519’ elliptic-curve cryptography”.

As we mentioned, AES-XTS is one of the modes of the AES block cipher. AES is based on a substitution-permutation network (SPN) structure, which combines substitution operations (SubBytes using the S-box), permutation operations (ShiftRows, MixColumns), and key mixing. By expanding s2n-bignum to include the AES instruction set architecture (ISA) found in Arm64 and x86_64 processors, specifications for the AES block cipher, and additional specifications for AES-XTS, we're paving the way for the same rigorous verification of more AES-based algorithms.

Developing and testing the specification

The SPN nature of AES and the modes that are based on it cannot be expressed using simple mathematical formulae — such as modular multiplication, which is fundamental to public-key cryptography — that can be innately understood by a theorem prover. They require writing descriptions of the steps for processing the data. This is why, before verifying the assembly, we needed confidence that our HOL Light specification accurately captured the IEEE standard.

We wrote the specification to mirror the standard's structure, using byte lists for input/output and 128-bit words for internal block operations. Then we developed conversions, HOL Light functions that we used to evaluate specifications with concrete inputs while generating proofs that the evaluations are mathematically correct.

We validated our specification by conducting unit tests that cover different AES-XTS encryption/decryption scenarios, exercising the processing of all blocks (using recursion) and ciphertext stealing.

These tests confirmed that our specification matched the IEEE standard before we tackled the more complex assembly verification. This two-phase approach — first ensuring that the specification is correct through testing, then formally verifying that the implementation matches the specification — gave us confidence we were proving the right thing.

The proof strategy

Our proofs are compositional, meaning they break the overall problem into subproblems that can be proved separately. Depending on the subproblem, the subproofs can be bounded — true only for a range of inputs — or unbounded.

For inputs with fewer than five (or six, in the case of decrypt) blocks, we wrote bounded proofs that exhaustively verify each case. For inputs with five (six, in the case of decrypt) or more blocks, we developed loop invariants — mathematical statements that remain true throughout loop execution — to prove correctness for arbitrarily long inputs. The loop invariants track three critical factors until the loop exit condition is met: register states at each iteration, the evolution of "tweaks" (which make each block's encryption unique), and memory contents as blocks are processed. For partial-block (tail) handling, we proved a separate theorem for ciphertext stealing that could be reused across all cases.

The top-level correctness theorem composes all proofs together, asserting the following statement:

If the program, inputs, output, and stack satisfy 
certain disjointness properties, and the input length len
satisfies 16 <= len <= 224, then given that the initial
machine state is set up with proper symbolic values, the value
stored in the output buffer must satisfy the AES-XTS 
specification after the whole program is executed.

Memory safety and constant-time proofs

Most recently, s2n-bignum was equipped with new functions and tactics for formally defining the constant-time and memory safety properties of assembly functions. With these resources, many assembly subroutines in s2n-bignum were verified to be constant time and memory safe, including top-level scalar-multiplication functions in elliptic curves, big-integer arithmetic for RSA, and the Arm implementation of the ML-KEM cryptography standard (the subject of a forthcoming blog post on Amazon Science). All assembly subroutines identified for use in AWS-LC as of October 2025 were formally verified to be constant time and memory safe.

We are exploring whether the new tactics can easily be used to verify assembly subroutines that have subsequently been added, such as AES-XTS. As we mentioned, AES-XTS has a remarkably complex control flow, which resulted in a long and involved functional-correctness proof. That complexity is also a challenge for safety proofs. The process is ongoing, but we have already proved safety properties for the ciphertext-stealing subroutines of the decryption and encryption algorithms.

These first proofs focused on crucial memory access procedures that are prone to buffer overread. Proofs for the remaining parts of the decryption and encryption algorithms can use the same methodology, where the constant-time and memory-safety proofs follow the same structure as the functional-correctness proofs but are simpler, since their proof goal is more focused.

Continuous assurance of correctness

We've integrated formal verification into s2n-bignum's continuous-integration (CI) workflow. This provides assurance that no changes to our AES-XTS implementation can be committed without successfully passing a formal proof of correctness. As part of CI, the CPU instruction modeling is validated through randomized testing against real hardware, "fuzzing out" inaccuracies to ensure our specifications are correct and the proofs hold in practice.

Furthermore, the proof guarantees correctness for all possible inputs, since they’re represented in the proof as symbols. This overcomes the typical shortcoming of coverage testing, which may cover all paths of the code but may not be able to cover all input values. For example, a constant-time code, like the one used here, is written without branching on secret values. Typically, then, secret values are incorporated into the operation through the use of masks derived from them. The same instructions are executed irrespective of the secret value. Hence, achieving line coverage is usually within the reach of a developer, but achieving value coverage is left to the formal verification of correctness.

This same methodology has enabled AWS to deploy optimized cryptographic implementations with mathematical guarantees of correctness while achieving significant performance improvements. This allows the developer and tools to further optimize the code freely without worrying about introducing bugs, since these will be automatically caught by the proof. Our experience with AES-XTS shows that proof-driven development of assembly code yields a control flow that is easier to understand, review, maintain, and optimize while never ceasing to be provably correct.

Research areas

Related content

US, CA, San Francisco
Amazon’s Frontier AI & Robotics (FAR) team is seeking a Member of Technical Staff to drive foundational research and build intelligent robotic systems from the ground up. In this role, you will operate at the intersection of innovative AI research and real-world robotics - conducting original research, publishing, and deploying your innovations into production systems at Amazon scale. We’re looking for researchers who think from first principles, push the boundaries of what’s possible, and take full ownership of turning breakthrough ideas into working systems. You will join the next revolution in robotics, where you'll work alongside world-renowned AI pioneers to push the boundaries of what's possible in robotic intelligence. As a Member of Technical Staff, you'll develop breakthrough foundation models that enable robots to perceive, understand, and interact with the physical world in unprecedented ways—with a strong emphasis on the hardware systems that bring these models to life. You'll drive independent research initiatives in areas such as locomotion, manipulation, motor control, actuator design, sim2real transfer, and multi-modal robot learning, designing novel frameworks that bridge state-of-the-art research with real-world hardware deployment at Amazon scale. In this role, you'll balance innovative technical exploration with hands-on hardware implementation, collaborating with mechanical, electrical, and controls engineering teams to ensure your models and algorithms perform robustly on physical robotic platforms in dynamic real-world environments. You'll have access to Amazon's computational resources and advanced robotics infrastructure—including high degree-of-freedom prototype platforms, custom actuators, and precision sensing systems—enabling you to tackle ambitious problems in areas like multi-modal robotic foundation models, motor-level control optimization, and efficient model architectures that scale across diverse robotic hardware. Key job responsibilities · Drive independent research initiatives across the full robotics stack, including robot co-design, manipulation mechanisms, innovative actuation and motor control strategies, state estimation, low-level control, system identification, reinforcement learning, and sim-to-real transfer, as well as foundation models for perception and manipulation · Lead full-stack robotics projects from conceptualization through hardware deployment, taking a system-level approach that integrates actuator dynamics, sensor feedback (force/torque, IMUs, encoders), and electromechanical constraints with algorithmic development · Develop and optimize control algorithms and sensing pipelines for physical robotic hardware, including motor characterization, actuator performance tuning, and robust sensor integration in production environments · Collaborate with hardware, mechanical, and electrical engineering teams to ensure seamless integration of learned models across the robotics stack—from embedded compute and communication buses to actuator-level control · Contribute to the team's technical strategy and help shape our approach to next-generation hardware-aware robotics challenges, including hardware-in-the-loop validation and prototype-to-deployment transitions A day in the life · Design and implement innovative systems and algorithms, leveraging our extensive computational and robotics hardware infrastructure to prototype and evaluate at scale · Collaborate with hardware and software engineers to solve complex technical challenges spanning motors, actuators, sensors, and learned control · Lead technical initiatives from conception to hardware deployment, working closely with robotics engineers and lab teams to integrate your solutions into physical robotic platforms · Participate in technical discussions and design reviews with team leaders, hardware engineers, and fellow scientists · Leverage our compute cluster and advanced robotics lab—including high-DoF prototype platforms and custom actuation systems—to rapidly prototype and validate new ideas · Transform theoretical insights into practical solutions that perform reliably on real-world robotic hardware About the team At Frontier AI & Robotics, we're not just advancing robotics – we're reimagining it from the ground up. Our team is building the future of intelligent robotics through ground breaking foundation models and end-to-end learned systems. We tackle some of the most challenging problems in AI and robotics, from developing sophisticated perception systems to creating adaptive manipulation strategies that work in complex, real-world scenarios. What sets us apart is our unique combination of ambitious research vision and practical impact. We leverage Amazon's massive computational infrastructure and rich real-world datasets to train and deploy state-of-the-art foundation models. Our work spans the full spectrum of robotics intelligence – from multimodal perception using images, videos, and sensor data, to sophisticated manipulation strategies that can handle diverse real-world scenarios. We're building systems that don't just work in the lab, but scale to meet the demands of Amazon's global operations. Join us if you're excited about pushing the boundaries of what's possible in robotics, working with world-class researchers, and seeing your innovations deployed at unprecedented scale.
GB, Virtual Location - Uk
Amazon Integrated Security is seeking a Principal Applied Scientist to lead research at the intersection of cognitive psychology, measurement science, and AI security. This role will pioneer new approaches to understanding and leveraging behavioral characteristics of autonomous AI systems in security contexts, turning threat actor use of AI into a structural advantage for defenders. Key job responsibilities Establish and lead a novel research program exploring cognitive patterns and biases in agentic AI systems used for offensive security purposes. Design and conduct experiments to understand black-box AI system behavior, develop measurement frameworks for previously unmeasurable security concepts, and translate research findings into actionable defensive strategies and tooling. Create feedback loops between offensive and defensive security teams, collaborate with threat intelligence experts and red team operations, and publish findings that advance both academic understanding and practical security outcomes. Apply measurement science expertise to quantify complex security concepts including risk, business impact, and trust. A day in the life Work closely with a Principal Technologist with AI and security expertise, senior security engineers with threat intelligence backgrounds, offensive AI security teams, and detection/response operations. Mentor junior scientists and engineers while maintaining strong connections with academic research communities.
US, NY, New York
We are seeking an Applied Scientist to lead the development of evaluation frameworks and data collection protocols for robotic capabilities. In this role, you will focus on designing how we measure, stress-test, and improve robot behavior across a wide range of real-world tasks. Your work will play a critical role in shaping how policies are validated and how high-quality datasets are generated to accelerate system performance. You will operate at the intersection of robotics, machine learning, and human-in-the-loop systems, building the infrastructure and methodologies that connect teleoperation, evaluation, and learning. This includes developing evaluation policies, defining task structures, and contributing to operator-facing interfaces that enable scalable and reliable data collection. The ideal candidate is highly experimental, systems-oriented, and comfortable working across software, robotics, and data pipelines, with a strong focus on turning ambiguous capability goals into measurable and actionable evaluation systems. Key job responsibilities - Design and implement evaluation frameworks to measure robot capabilities across structured tasks, edge cases, and real-world scenarios - Develop task definitions, success criteria, and benchmarking methodologies that enable consistent and reproducible evaluation of policies - Create and refine data collection protocols that generate high-quality, task-relevant datasets aligned with model development needs - Build and iterate on teleoperation workflows and operator interfaces to support efficient, reliable, and scalable data collection - Analyze evaluation results and collected data to identify performance gaps, failure modes, and opportunities for targeted data collection - Collaborate with engineering teams to integrate evaluation tooling, logging systems, and data pipelines into the broader robotics stack - Stay current with advances in robotics, evaluation methodologies, and human-in-the-loop learning to continuously improve internal approaches - Lead technical projects from conception through production deployment - Mentor junior scientists and engineers About the team Fauna Robotics, an Amazon company, is building capable, safe, and genuinely delightful robots for everyday life. Our goal is simple: make robots people actually want to live and interact with in everyday human spaces. We believe that future won’t arrive until building for robotics becomes far more accessible. Today, too much effort is spent reinventing the fundamentals. We’re changing that by developing tightly integrated hardware and software systems that make it faster, safer, and more intuitive to create real-world robotic products. Our work spans the full stack: mechanical design, control systems, dynamic modeling, and intelligent software. The focus is not just functionality, but experience. We’re building robots that feel responsive, expressive, and genuinely useful. At Fauna, you’ll work at the frontier of this space, helping define how robots move, manipulate, and interact with people in natural environments. It’s an opportunity to solve hard problems across hardware and software with a team focused on making robotics accessible and joyful to build. If you care about making robotics real for everyone and building systems that are as delightful as they are capable, we’re interested in hearing from you.
US, NY, New York
We are seeking a Robotics/AI Motor Control Scientist to develop cutting-edge machine learning algorithms for motor control systems in robots. In this role, you will focus on creating and optimizing intelligent motor control strategies to enable robots to perform complex, whole-body tasks. Your contributions will be essential in advancing robotics by enabling fluid, reliable, and safe interactions between robots and their environments. Key job responsibilities - Develop controllers that leverage reinforcement learning, imitation learning, or other advanced AI techniques to achieve natural, robust, and adaptive motor behaviors - Collaborate with multi-disciplinary teams to integrate motor control systems with robotic hardware, ensuring alignment with real-world constraints such as actuator dynamics and energy efficiency - Use simulation and real-world testing to refine and validate control algorithms - Stay updated on advancements in robotics, AI, and control systems to apply advanced techniques to robotic motion challenges - Lead technical projects from conception through production deployment - Mentor junior scientists and engineers - Bridge research initiatives with practical engineering implementation About the team Fauna Robotics, an Amazon company, is building capable, safe, and genuinely delightful robots for everyday life. Our goal is simple: make robots people actually want to live and interact with in everyday human spaces. We believe that future won’t arrive until building for robotics becomes far more accessible. Today, too much effort is spent reinventing the fundamentals. We’re changing that by developing tightly integrated hardware and software systems that make it faster, safer, and more intuitive to create real-world robotic products. Our work spans the full stack: mechanical design, control systems, dynamic modeling, and intelligent software. The focus is not just functionality, but experience. We’re building robots that feel responsive, expressive, and genuinely useful. At Fauna, you’ll work at the frontier of this space, helping define how robots move, manipulate, and interact with people in natural environments. It’s an opportunity to solve hard problems across hardware and software with a team focused on making robotics accessible and joyful to build. If you care about making robotics real for everyone and building systems that are as delightful as they are capable, we’re interested in hearing from you. an opportunity to solve hard problems across hardware and software with a team focused on making robotics accessible and joyful to build. If you care about making robotics real for everyone and building systems that are as delightful as they are capable, we’re interested in hearing from you.
AU, VIC, Melbourne
Are you excited about leveraging state-of-the-art Computer Vision algorithms and large datasets to solve real-world problems? Join Amazon as an Applied Scientist Intern and be at the forefront of AI innovation! As an Applied Scientist Intern, you'll work in a fast-paced, cross-disciplinary team of pioneering researchers. You'll tackle complex problems, developing solutions that either build on existing academic and industrial research or stem from your own innovative thinking. Your work may even find its way into customer-facing products, making a real-world impact. Key job responsibilities - Develop novel solutions and build prototypes - Work on complex problems in Computer Vision and Machine Learning - Contribute to research that could significantly impact Amazon's operations - Collaborate with a diverse team of experts in a fast-paced environment - Collaborate with scientists on writing and submitting papers to Tier-1 conferences (e.g., CVPR, ICCV, NeurIPS, ICML) - Present your research findings to both technical and non-technical audiences Key Opportunities: - Collaborate with leading machine learning researchers - Access innovative tools and hardware (large GPU clusters) - Address challenges at an unparalleled scale - Become a disruptor, innovator, and problem solver in the field of computer vision - Potentially deliver solutions to production in customer-facing applications - Opportunities to become an FTE after the internship Join us in shaping the future of AI at Amazon. Apply now and turn your research into real-world solutions!
US, WA, Seattle
The Alexa for Shopping team is seeking a customer-obsessed senior economist to own and drive analytics strategy for GenAI-powered Shopping experiences. This role will partner closely with senior leaders to deliver high-quality insights that inform executive decision-making for the AI shopping assistant, Rufus. The successful candidate will demonstrate strong attention to detail, excellent written and verbal communication, and the ability to influence across organizations. In this role, you will mentor and set the bar for data science, economics, and engineering partners by establishing best practices for understanding customer behavior in AI-driven shopping experiences. You will invent and scale metrics that measure customer adoption and habituation, and build agentic, automated analytical workflows that enable fast, repeatable deep dives. This position will play a critical role in shaping product roadmap and investment decisions in a rapidly evolving GenAI space. The ideal candidate will operate effectively in ambiguous environments, exercise strong business judgment on high-impact, one-way door decisions, and continuously raise the bar for analytical rigor and operational excellence. You will work cross-functionally with product, engineering, and economics partners to deliver results for customers Key job responsibilities - Own the development of customer and shopping-mission cohorts to understand behavior with and without Rufus engagement across the end-to-end shopping journey. - Identify which Rufus query types and interaction patterns drive the most customer value for specific customer cohorts and shopping missions. - Build predictive models to estimate customer re-engagement and long-term adoption of Rufus based on interaction quality and downstream shopping outcomes. - Invent, operationalize, and publish scalable metrics and dashboards that surface actionable insights, enabling data-driven product growth and executive decision-making. - Partner closely with Product, Engineering, and Economics teams to translate analytical insights into roadmap priorities and customer-focused improvements. About the team The Alexa for Shopping economics team focuses on understanding how GenAI-powered shopping tools are transforming customer behavior across the shopping lifecycle - from inspiration and problem-solving, to product research, selection, purchase, and post-purchase support. We build the foundational measurement frameworks that enable teams to evaluate performance, identify what Rufus experiences resonate most with customers, and uncover opportunities for improvement. Our work directly influences customer-centric product roadmap decisions and helps scale impactful, high-quality AI shopping experiences
US, CA, San Francisco
Amazon’s Frontier AI & Robotics (FAR) team is seeking a Member of Technical Staff to drive foundational research and build intelligent robotic systems from the ground up. In this role, you will operate at the intersection of cutting-edge AI research and real-world robotics - conducting original research, publishing, and deploying your innovations into production systems at Amazon scale. We’re looking for researchers who think from first principles, push the boundaries of what’s possible, and take full ownership of turning breakthrough ideas into working systems.  You will join the next revolution in robotics, where you'll work alongside world-renowned AI pioneers to push the boundaries of what's possible in robotic intelligence. As a Member of Technical Staff, you'll be at the forefront of developing breakthrough foundation models and full-stack robotics systems that enable robots to perceive, understand, and interact with the world in unprecedented ways. You'll drive technical excellence and independent research initiatives in areas such as locomotion, manipulation, perception, sim2real transfer, multi-modal, multi-task robot learning, designing novel frameworks that bridge the gap between state-of-the-art research and real-world deployment at Amazon scale. In this role, you'll balance innovative technical exploration with practical implementation, collaborating with platform teams to ensure your models and algorithms perform robustly in dynamic real-world environments. You’ll have the freedom to pursue ambitious research directions while leveraging Amazon’s vast computational resources to tackle ambiguous problems in areas like very large multi-modal robotic foundation models and efficient, promptable model architectures that can scale across diverse robotic applications. Key job responsibilities - Drive independent research initiatives across the robotics stack, including robot co-design, dexterous manipulation mechanisms, innovative actuation strategies, state estimation, low-level control, system identification, reinforcement learning, sim-to-real transfer, as well as foundation models focusing on breakthrough approaches in perception, and manipulation, for example open-vocabulary panoptic scene understanding, scaling up multi-modal LLMs, sim2real/real2sim techniques, end-to-end vision-language-action models, efficient model inference, video tokenization - Design and implement novel deep learning architectures that push the boundaries of what robots can understand and accomplish - Guide technical direction for full-stack robotics projects from conceptualization through deployment, taking a system-level approach that integrates hardware considerations with algorithmic development, ensuring robust performance in production environments - Collaborate with platform and hardware teams to ensure seamless integration across the entire robotics stack, optimizing and scaling models for real-world applications - Contribute to team's technical decisions and influence implementation strategies to help shape our approach to next-generation robotics challenges - Mentor fellow researchers while maintaining solid individual technical contributions A day in the life - Design and implement novel foundation model architectures and innovative systems and algorithms, leveraging our extensive infrastructure to prototype and evaluate at scale - Collaborate with our world-class research team to solve complex technical challenges across the full robotics stack - Lead focused technical initiatives from conception through deployment, ensuring successful integration with production systems - Drive technical discussions and brainstorming sessions with team leaders, fellow researchers and key stakeholders - Conduct experiments and prototype new ideas using our massive compute cluster and extensive robotics infrastructure - Transform theoretical insights into practical solutions that can handle the complexities of real-world robotics applications About the team At Frontier AI & Robotics, we're not just advancing robotics – we're reimagining it from the ground up. Our team is building the future of intelligent robotics through innovative foundation models and end-to-end learned systems. We tackle some of the most challenging problems in AI and robotics, from developing sophisticated perception systems to creating adaptive manipulation strategies that work in complex, real-world scenarios. What sets us apart is our unique combination of ambitious research vision and practical impact. We leverage Amazon's massive computational infrastructure and rich real-world datasets to train and deploy state-of-the-art foundation models. Our work spans the full spectrum of robotics intelligence – from multimodal perception using images, videos, and sensor data, to sophisticated manipulation strategies that can handle diverse real-world scenarios. We're building systems that don't just work in the lab, but scale to meet the demands of Amazon's global operations. Join us if you're excited about pushing the boundaries of what's possible in robotics, working with world-class researchers, and seeing your innovations deployed at unprecedented scale.
US, NY, New York
We are seeking a Sr. Applied Scientist to develop cutting-edge machine learning algorithms for motor control systems in robots. In this role, you will focus on creating and optimizing intelligent motor control strategies to enable robots to perform complex, whole-body tasks. Your contributions will be essential in advancing robotics by enabling fluid, reliable, and safe interactions between robots and their environments. Key job responsibilities - Develop controllers that leverage reinforcement learning, imitation learning, or other advanced AI techniques to achieve natural, robust, and adaptive motor behaviors - Collaborate with multi-disciplinary teams to integrate motor control systems with robotic hardware, ensuring alignment with real-world constraints such as actuator dynamics and energy efficiency - Use simulation and real-world testing to refine and validate control algorithms - Stay updated on advancements in robotics, AI, and control systems to apply advanced techniques to robotic motion challenges - Lead technical projects from conception through production deployment - Mentor junior scientists and engineers - Bridge research initiatives with practical engineering implementation About the team Fauna Robotics, an Amazon company, is building capable, safe, and genuinely delightful robots for everyday life. Our goal is simple: make robots people actually want to live and interact with in everyday human spaces. We believe that future won’t arrive until building for robotics becomes far more accessible. Today, too much effort is spent reinventing the fundamentals. We’re changing that by developing tightly integrated hardware and software systems that make it faster, safer, and more intuitive to create real-world robotic products. Our work spans the full stack: mechanical design, control systems, dynamic modeling, and intelligent software. The focus is not just functionality, but experience. We’re building robots that feel responsive, expressive, and genuinely useful. At Fauna, you’ll work at the frontier of this space, helping define how robots move, manipulate, and interact with people in natural environments. It’s an opportunity to solve hard problems across hardware and software with a team focused on making robotics accessible and joyful to build. If you care about making robotics real for everyone and building systems that are as delightful as they are capable, we’re interested in hearing from you. an opportunity to solve hard problems across hardware and software with a team focused on making robotics accessible and joyful to build. If you care about making robotics real for everyone and building systems that are as delightful as they are capable, we’re interested in hearing from you.
US, CA, San Francisco
Amazon’s Frontier AI & Robotics (FAR) team is seeking a Member of Technical Staff to drive foundational research and build intelligent robotic systems from the ground up. In this role, you will operate at the intersection of cutting-edge AI research and real-world robotics - conducting original research, publishing, and deploying your innovations into production systems at Amazon scale. We’re looking for researchers who think from first principles, push the boundaries of what’s possible, and take full ownership of turning breakthrough ideas into working systems.  You will join the next revolution in robotics, where you'll work alongside world-renowned AI pioneers to push the boundaries of what's possible in robotic intelligence. As a Member of Technical Staff, you'll be at the forefront of developing breakthrough foundation models and full-stack robotics systems that enable robots to perceive, understand, and interact with the world in unprecedented ways. You'll drive technical excellence and independent research initiatives in areas such as locomotion, manipulation, perception, sim2real transfer, multi-modal, multi-task robot learning, designing novel frameworks that bridge the gap between state-of-the-art research and real-world deployment at Amazon scale. In this role, you'll balance innovative technical exploration with practical implementation, collaborating with platform teams to ensure your models and algorithms perform robustly in dynamic real-world environments. You’ll have the freedom to pursue ambitious research directions while leveraging Amazon’s vast computational resources to tackle ambiguous problems in areas like very large multi-modal robotic foundation models and efficient, promptable model architectures that can scale across diverse robotic applications. Key job responsibilities - Drive independent research initiatives across the robotics stack, driving breakthrough approaches through hands-on research and development in areas including robot co-design, dexterous manipulation mechanisms, innovative actuation strategies, state estimation, low-level control, system identification, reinforcement learning, sim-to-real transfer, as well as foundation models focusing on breakthrough approaches in perception, and manipulation. - Lead and Guide technical direction for full-stack robotics projects from conceptualization through deployment, taking a system-level approach that integrates hardware considerations with algorithmic development - Develop and optimize control algorithms and sensing pipelines that enable robust performance in production environments - Collaborate with platform and hardware teams to ensure seamless integration across the entire robotics stack, optimizing and scaling models for real-world applications - Contribute to team's technical decisions and influence implementation strategies to help shape our approach to next-generation robotics challenges - Mentor fellow researchers while maintaining solid individual technical contributions A day in the life - Design and implement novel foundation model architectures and innovative systems and algorithms, leveraging our extensive infrastructure to prototype and evaluate at scale - Collaborate with our world-class research team to solve complex technical challenges across the full robotics stack - Lead focused technical initiatives from conception through deployment, ensuring successful integration with production systems - Drive technical discussions and brainstorming sessions with team leaders, fellow researchers and key stakeholders - Conduct experiments and prototype new ideas using our massive compute cluster and extensive robotics infrastructure - Transform theoretical insights into practical solutions that can handle the complexities of real-world robotics applications About the team At Frontier AI & Robotics, we're not just advancing robotics – we're reimagining it from the ground up. Our team is building the future of intelligent robotics through innovative foundation models and end-to-end learned systems. We tackle some of the most challenging problems in AI and robotics, from developing sophisticated perception systems to creating adaptive manipulation strategies that work in complex, real-world scenarios. What sets us apart is our unique combination of ambitious research vision and practical impact. We leverage Amazon's massive computational infrastructure and rich real-world datasets to train and deploy state-of-the-art foundation models. Our work spans the full spectrum of robotics intelligence – from multimodal perception using images, videos, and sensor data, to sophisticated manipulation strategies that can handle diverse real-world scenarios. We're building systems that don't just work in the lab, but scale to meet the demands of Amazon's global operations. Join us if you're excited about pushing the boundaries of what's possible in robotics, working with world-class researchers, and seeing your innovations deployed at unprecedented scale.
IN, KA, Bengaluru
Amazon is looking for a passionate, talented, and inventive Applied Scientists with machine learning background to help build industry-leading Speech and Language technology. Our mission is to provide a delightful experience to Amazon’s customers by pushing the envelope in Automatic Speech Recognition (ASR), Natural Language Understanding (NLU), Machine Learning (ML) and Computer Vision (CV). Key job responsibilities Amazon is looking for a passionate, talented, and inventive Applied Scientists with machine learning background to help build industry-leading Speech and Language technology. Our mission is to provide a delightful experience to Amazon’s customers by pushing the envelope in Automatic Speech Recognition (ASR), Natural Language Understanding (NLU), Machine Learning (ML) and Computer Vision (CV). As part of our AI team in Amazon AWS, you will work alongside internationally recognized experts to develop novel algorithms and modeling techniques to advance the state-of-the-art in human language technology. Your work will directly impact millions of our customers in the form of products and services that make use of speech and language technology. You will gain hands on experience with Amazon’s heterogeneous speech, text, and structured data sources, and large-scale computing resources to accelerate advances in spoken language understanding. We are hiring in all areas of human language technology: ASR, MT, NLU, text-to-speech (TTS), and Dialog Management, in addition to Computer Vision. We are also looking for talents with experiences/expertise in building large-scale, high-performing systems. A day in the life 0