Social
Conversational AI

A quick guide to Amazon’s 50-plus ICASSP papers 2022

Topics range from the predictable, such as speech recognition and signal processing, to time series forecasting and personalization.

By Staff writer
May 10, 2022
7 min read
Share

Amazon researchers have more than 50 papers at this year’s International Conference on Acoustics, Speech, and Signal Processing (ICASSP). A plurality of them are on automatic speech recognition and related topics, such as keyword spotting and speaker identification. But others range farther afield, to topics such as computer vision and federated learning.

ICASSP-2022-Header.png
This year's ICASSP includes a virtual component, from May 7 to 13, and an in-person component in Singapore, May 22 to 27.

Acoustic-event detection

Federated self-supervised learning for acoustic event classification
Meng Feng, Chieh-Chi Kao, Qingming Tang, Ming Sun, Viktor Rozgic, Spyros Matsoukas, Chao Wang

Improved representation learning for acoustic event classification using tree-structured ontology
Arman Zharmagambetov, Qingming Tang, Chieh-Chi Kao, Qin Zhang, Ming Sun, Viktor Rozgic, Jasha Droppo, Chao Wang

WikiTAG: Wikipedia-based knowledge embeddings towards improved acoustic event classification
Qin Zhang, Qingming Tang, Chieh-Chi Kao, Ming Sun, Yang Liu, Chao Wang

Automatic speech recognition

A likelihood ratio-based domain adaptation method for end-to-end models
Chhavi Choudhury, Ankur Gandhe, Xiaohan Ding, Ivan Bulyko

Being greedy does not hurt: Sampling strategies for end-to-end speech recognition
Jahn Heymann, Egor Lakomkin, Leif RādellJahn Heymann, Egor Lakomkin, Leif RādelJahn Heymann, Egor Lakomkin, Leif RādelJahn Heymann, Egor Lakomkin, Leif Rādel

Caching networks: Capitalizing on common speech for ASR
Anastasios Alexandridis, Grant P. Strimel, Ariya Rastrow, Pavel Kveton, Jon Webb, Maurizio Omologo, Siegfried Kunzmann, Athanasios Mouchtaris

Lattice attention.png
In "LATTENTION: Lattice attention in ASR rescoring", Amazon researchers show that applying an attention mechanism (colored grid) to a lattice encoding multiple automatic-speech-recognition (ASR) hypotheses improves ASR performance.

Contextual adapters for personalized speech recognition in neural transducers
Kanthashree Mysore Sathyendra, Thejaswi Muniyappa, Feng-Ju Chang, Jing Liu, Jinru Su, Grant P. Strimel, Athanasios Mouchtaris, Siegfried Kunzmann

LATTENTION: Lattice attention in ASR rescoring
Prabhat Pandey, Sergio Duarte Torres, Ali Orkan Bayer, Ankur Gandhe, Volker Leutnant

Listen, know and spell: Knowledge-infused subword modeling for improving ASR performance of out-of-vocabulary (OOV) named entities
Nilaksh Das, Monica Sunkara, Dhanush Bekal, Duen Horng Chau, Sravan Bodapati, Katrin Kirchhoff

KG ASR rescoring.png
In "Listen, know and spell: Knowledge-infused subword modeling for improving ASR performance of OOV named entities", Amazon researchers show how to improve automatic speech recognition by incorporating information from knowledge graphs into the processing pipeline.

Mitigating closed-model adversarial examples with Bayesian neural modeling for enhanced end-to-end speech recognition
Chao-Han Huck Yang, Zeeshan Ahmed, Yile Gu, Joseph Szurley, Roger Ren, Linda Liu, Andreas Stolcke, Ivan Bulyko

Multi-modal pre-training for automated speech recognition
David M. Chan, Shalini Ghosh, Debmalya Chakrabarty, Björn Hoffmeister

Multiturn encoder.png
The model used in "Multi-turn RNN-T for streaming recognition of multi-party speech" to disentangle overlapping speech in multi-party automatic speech recognition.

Multi-turn RNN-T for streaming recognition of multi-party speech
Ilya Sklyar, Anna Piunova, Xianrui Zheng, Yulan Liu

RescoreBERT: Discriminative speech recognition rescoring with BERT
Liyan Xu, Yile Gu, Jari Kolehmainen, Haidar Khan, Ankur Gandhe, Ariya Rastrow, Andreas Stolcke, Ivan Bulyko

USTED: Improving ASR with a unified speech and text encoder-decoder
Bolaji Yusuf, Ankur Gandhe, Alex Sokolov

VADOI: Voice-activity-detection overlapping inference for end-to-end long-form speech recognition
Jinhan Wang, Xiaosu Tong, Jinxi Guo, Di He, Roland Maas

Computer vision

ASD-transformer: Efficient active speaker detection using self and multimodal transformers
Gourav Datta, Tyler Etchart, Vivek Yadav, Varsha Hedau, Pradeep Natarajan, Shih-Fu Chang

Dynamically pruning SegFormer for efficient semantic segmentation
Haoli Bai, Hongda Mao, Dinesh Nair

Enhancing contrastive learning with temporal cognizance for audio-visual representation generation
Chandrashekhar Lavania, Shiva Sundaram, Sundararajan Srinivasan, Katrin Kirchhoff

Few-shot gaze estimation with model offset predictors
Jiawei Ma, Xu Zhang, Yue Wu, Varsha Hedau, Shih-Fu Chang

Visual representation learning with self-supervised attention for low-label high-data regime
Prarthana Bhattacharyya, Chenge Li, Xiaonan Zhao, István Fehérvári, Jason Sun

Federated learning

Federated learning challenges and opportunities: An outlook
Jie Ding, Eric Tramel, Anit Kumar Sahu, Shuang Wu, Salman Avestimehr, Tao Zhang

FL framework.png
The federated-learning scenario considered in "Federated learning challenges and opportunities: An outlook".

Learnings from federated learning in the real world
Christophe Dupuy, Tanya G. Roosta, Leo Long, Clement Chung, Rahul Gupta, Salman Avestimehr

Information retrieval

Contrastive knowledge graph attention network for request-based recipe recommendation
Xiyao Ma, Zheng Gao, Qian Hu, Mohamed Abdelhady

Keyword spotting

Unified speculation, detection, and verification keyword spotting
Geng-shen Fu, Thibaud Senechal, Aaron Challenner, Tao Zhang

Machine translation

Isometric MT: Neural machine translation for automatic dubbing
Surafel Melaku Lakew, Yogesh Virkar, Prashant Mathur, Marcello Federico

Natural-language understanding

ADVIN: Automatically discovering novel domains and intents from user text utterances
Nikhita Vedula, Rahul Gupta, Aman Alok, Mukund Sridhar, Shankar Ananthakrishnan

An efficient DP-SGD mechanism for large scale NLU models
Christophe Dupuy, Radhika Arava, Rahul Gupta, Anna Rumshisky

Paralinguistics

Confidence estimation for speech emotion recognition based on the relationship between emotion categories and primitives
Yang Li, Constantinos Papayiannis, Viktor Rozgic, Elizabeth Shriberg, Chao Wang

Multi-lingual multi-task speech emotion recognition using wav2vec 2.0
Mayank Sharma

Representation learning through cross-modal conditional teacher-student training for speech emotion recognition
Sundararajan Srinivasan, Zhaocheng Huang, Katrin Kirchhoff

Sentiment-aware automatic speech recognition pre-training for enhanced speech emotion recognition
Ayoub Ghriss, Bo Yang, Viktor Rozgic, Elizabeth Shriberg, Chao Wang

Personalization

Incremental user embedding modeling for personalized text classification
Ruixue Lian, Che-Wei Huang, Yuqing Tang, Qilong Gu, Chengyuan Ma, Chenlei (Edward) Guo

Signal processing

Deep adaptive AEC: Hybrid of deep learning and adaptive acoustic echo cancellation
Hao Zhang, Srivatsan Kandadai, Harsha Rao, Minje Kim, Tarun Pruthi, Trausti Kristjansson

Improved singing voice separation with chromagram-based pitch-aware remixing
Siyuan Yuan, Zhepei Wang, Umut Isik, Ritwik Giri, Jean-Marc Valin, Michael M. Goodwin, Arvindh Krishnaswamy

Sparse recovery of acoustic waves
Mohamed Mansour

Upmixing via style transfer: A variational autoencoder for disentangling spatial images and musical content
Haici Yang, Sanna Wager, Spencer Russell, Mike Luo, Minje Kim, Wontak Kim

Sound source localization

End-to-end Alexa device arbitration
Jarred Barber, Yifeng Fan, Tao Zhang

Speaker diarization/identification/verification

ASR-aware end-to-end neural diarization
Aparna Khare, Eunjung Han, Yuguang Yang, Andreas Stolcke

Improving fairness in speaker verification via group-adapted fusion network
Hua Shen, Yuguang Yang, Guoli Sun, Ryan Langman, Eunjung Han, Jasha Droppo, Andreas Stolcke

OpenFEAT: Improving speaker identification by open-set few-shot embedding adaptation with Transformer
Kishan K C, Zhenning Tan, Long Chen, Minho Jin, Eunjung Han, Andreas Stolcke, Chul Lee

Self-supervised speaker recognition training using human-machine dialogues
Metehan Cekic, Ruirui Li, Zeya Chen, Yuguang Yang, Andreas Stolcke, Upamanyu Madhow

Self-supervised speaker verification with simple Siamese network and self-supervised regularization
Mufan Sang, Haoqi Li, Fang Liu, Andrew O. Arnold, Li Wan

Spoken-language understanding

A neural prosody encoder for end-to-end dialogue act classification
Kai Wei, Dillon Knox, Martin Radfar, Thanh Tran, Markus Mueller, Grant P. Strimel, Nathan Susanj, Athanasios Mouchtaris, Maurizio Omologo

Multi-task RNN-T with semantic decoder for streamable spoken language understanding
Xuandi Fu, Feng-Ju Chang, Martin Radfar, Kai Wei, Jing Liu, Grant P. Strimel, Kanthashree Mysore Sathyendra

Tie your embeddings down: Cross-modal latent spaces for end-to-end spoken language understanding
Bhuvan Agrawal, Markus Mueller, Samridhi Choudhary, Martin Radfar, Athanasios Mouchtaris, Ross McGowan, Nathan Susanj, Siegfried Kunzmann

TINYS2I: A small-footprint utterance classification model with contextual support for on-device SLU
Anastasios Alexandridis, Kanthashree Mysore Sathyendra, Grant P. Strimel, Pavel Kveton, Jon Webb, Athanasios Mouchtaris

Text-to-speech

Cross-speaker style transfer for text-to-speech using data augmentation
Manuel Sam Ribeiro, Julian Roth, Giulia Comini, Goeric Huybrechts, Adam Gabrys, Jaime Lorenzo-Trueba

Distribution augmentation for low-resource expressive text-to-speech
Mateusz Lajszczak, Animesh Prasad, Arent van Korlaar, Bajibabu Bollepalli, Antonio Bonafonte, Arnaud Joly, Marco Nicolis, Alexis Moinet, Thomas Drugman, Trevor Wood, Elena Sokolova

Duration modeling of neural TTS for automatic dubbing
Johanes Effendi, Yogesh Virkar, Roberto Barra-Chicote, Marcello Federico

Neural speech synthesis on a shoestring: Improving the efficiency of LPCNET
Jean-Marc Valin, Umut Isik, Paris Smaragdis, Arvindh Krishnaswamy

Text-free non-parallel many-to-many voice conversion using normalising flows
Thomas Merritt, Abdelhamid Ezzerg, Piotr Biliński, Magdalena Proszewska, Kamil Pokora, Roberto Barra-Chicote, Daniel Korzekwa

VoiceFilter: Few-shot text-to-speech speaker adaptation using voice conversion as a post-processing module
Adam Gabrys, Goeric Huybrechts, Manuel Sam Ribeiro, Chung-Ming Chien, Julian Roth, Giulia Comini, Roberto Barra-Chicote, Bartek Perz, Jaime Lorenzo-Trueba

Time series forecasting

Robust nonparametric distribution forecast with backtest-based bootstrap and adaptive residual selection
Longshaokan Marshall Wang, Lingda Wang, Mina Georgieva, Paulo Machado, Abinaya Ulagappa, Safwan Ahmed, Yan Lu, Arjun Bakshi, Farhad Ghassemi

Research areas
Tags
About the Author
Staff writer

Related content

Work with us

See more jobs See more jobs
Data Scientist II, FinAuto
IN, KA, Bengaluru
Interested to build the next generation Financial systems that can handle billions of dollars in transactions? Interested to build highly scalable next generation systems that could utilize Amazon Cloud? Massive data volume + complex business rules in a highly distributed and service oriented architecture, a world class information collection and delivery challenge. Our challenge is to deliver the software systems which accurately capture, process, and report on the huge volume of financial transactions that are generated each day as millions of customers make purchases, as thousands of Vendors and Partners are paid, as inventory moves in and out of warehouses, as commissions are calculated, and as taxes are collected in hundreds of jurisdictions worldwide. Key job responsibilities • Understand the business and discover actionable insights from large volumes of data through application of machine learning, statistics or causal inference. • Analyse and extract relevant information from large amounts of Amazon’s historical transactions data to help automate and optimize key processes • Research, develop and implement novel machine learning and statistical approaches for anomaly, theft, fraud, abusive and wasteful transactions detection. • Use machine learning and analytical techniques to create scalable solutions for business problems. • Identify new areas where machine learning can be applied for solving business problems. • Partner with developers and business teams to put your models in production. • Mentor other scientists and engineers in the use of ML techniques. A day in the life • Understand the business and discover actionable insights from large volumes of data through application of machine learning, statistics or causal inference. • Analyse and extract relevant information from large amounts of Amazon’s historical transactions data to help automate and optimize key processes • Research, develop and implement novel machine learning and statistical approaches for anomaly, theft, fraud, abusive and wasteful transactions detection. • Use machine learning and analytical techniques to create scalable solutions for business problems. • Identify new areas where machine learning can be applied for solving business problems. • Partner with developers and business teams to put your models in production. • Mentor other scientists and engineers in the use of ML techniques. About the team The FinAuto TFAW(theft, fraud, abuse, waste) team is part of FGBS Org and focuses on building applications utilizing machine learning models to identify and prevent theft, fraud, abusive and wasteful(TFAW) financial transactions across Amazon. Our mission is to prevent every single TFAW transaction. As a Machine Learning Scientist in the team, you will be driving the TFAW Sciences roadmap, conduct research to develop state-of-the-art solutions through a combination of data mining, statistical and machine learning techniques, and coordinate with Engineering team to put these models into production. You will need to collaborate effectively with internal stakeholders, cross-functional teams to solve problems, create operational efficiencies, and deliver successfully against high organizational standards.
Applied Scientist, FinAuto
IN, KA, Bengaluru
Interested to build the next generation Financial systems that can handle billions of dollars in transactions? Interested to build highly scalable next generation systems that could utilize Amazon Cloud? Massive data volume + complex business rules in a highly distributed and service oriented architecture, a world class information collection and delivery challenge. Our challenge is to deliver the software systems which accurately capture, process, and report on the huge volume of financial transactions that are generated each day as millions of customers make purchases, as thousands of Vendors and Partners are paid, as inventory moves in and out of warehouses, as commissions are calculated, and as taxes are collected in hundreds of jurisdictions worldwide. Key job responsibilities • Understand the business and discover actionable insights from large volumes of data through application of machine learning, statistics or causal inference. • Analyse and extract relevant information from large amounts of Amazon’s historical transactions data to help automate and optimize key processes • Research, develop and implement novel machine learning and statistical approaches for anomaly, theft, fraud, abusive and wasteful transactions detection. • Use machine learning and analytical techniques to create scalable solutions for business problems. • Identify new areas where machine learning can be applied for solving business problems. • Partner with developers and business teams to put your models in production. • Mentor other scientists and engineers in the use of ML techniques. A day in the life • Understand the business and discover actionable insights from large volumes of data through application of machine learning, statistics or causal inference. • Analyse and extract relevant information from large amounts of Amazon’s historical transactions data to help automate and optimize key processes • Research, develop and implement novel machine learning and statistical approaches for anomaly, theft, fraud, abusive and wasteful transactions detection. • Use machine learning and analytical techniques to create scalable solutions for business problems. • Identify new areas where machine learning can be applied for solving business problems. • Partner with developers and business teams to put your models in production. • Mentor other scientists and engineers in the use of ML techniques. About the team The FinAuto TFAW(theft, fraud, abuse, waste) team is part of FGBS Org and focuses on building applications utilizing machine learning models to identify and prevent theft, fraud, abusive and wasteful(TFAW) financial transactions across Amazon. Our mission is to prevent every single TFAW transaction. As a Machine Learning Scientist in the team, you will be driving the TFAW Sciences roadmap, conduct research to develop state-of-the-art solutions through a combination of data mining, statistical and machine learning techniques, and coordinate with Engineering team to put these models into production. You will need to collaborate effectively with internal stakeholders, cross-functional teams to solve problems, create operational efficiencies, and deliver successfully against high organizational standards.
Applied Scientist II, HST Health Evaluation
IN, KA, Bengaluru
Amazon Health Services (One Medical) About Us: At Health AI, we're revolutionizing healthcare delivery through innovative AI-enabled solutions. As part of Amazon Health Services and One Medical, we're on a mission to make quality healthcare more accessible while improving patient outcomes. Our work directly impacts millions of lives by empowering patients and enabling healthcare providers to deliver more meaningful care. Role Overview: We're seeking an Applied Scientist to join our dynamic team in building state of the art AI/ML solutions for healthcare. This role offers a unique opportunity to work at the intersection of artificial intelligence and healthcare, developing solutions that will shape the future of medical services delivery. Key job responsibilities • Lead end-to-end development of AI/ML solutions for Amazon Health organization, including Amazon Pharmacy and One Medical • Research, design, and implement state-of-the-art machine learning models, with a focus on Large Language Models (LLMs) and Visual Language Models (VLMs) • Optimize and fine-tune models for production deployment, including model distillation for improved latency • Drive scientific innovation while maintaining a strong focus on practical business outcomes • Collaborate with cross-functional teams to translate complex technical solutions into tangible customer benefits • Contribute to the broader Amazon Health scientific community and help shape our technical roadmap
Mixed Signal Design Engineer, Amazon Center for Quantum Computing
US, CA, Pasadena
The Amazon Center for Quantum Computing in Pasadena, CA, is looking to hire an Applied Scientist specializing in Mixed-Signal Design. Working alongside other scientists and engineers, you will design and validate hardware performing the control and readout functions for AWS quantum processors. Candidates must have a solid background in mixed-signal design at the printed circuit board (PCB) level. Working effectively within a cross-functional team environment is critical. The ideal candidate will have demonstrated the capability to contribute to all phases of product life cycle development, from requirements gathering to verification. Diverse Experiences Amazon values diverse experiences. Even if you do not meet all of the preferred qualifications and skills listed in the job description, we encourage candidates to apply. If your career is just starting, hasn’t followed a traditional path, or includes alternative experiences, don’t let it stop you from applying. Work/Life Balance We value work-life harmony. Achieving success at work should never come at the expense of sacrifices at home, which is why we strive for flexibility as part of our working culture. When we feel supported in the workplace and at home, there’s nothing we can’t achieve in the cloud. Inclusive Team Culture Here at Amazon, it’s in our nature to learn and be curious. Our employee-led affinity groups foster a culture of inclusion that empower us to be proud of our differences. Ongoing events and learning experiences, including our Conversations on Race and Ethnicity (CORE) and AmazeCon (gender diversity) conferences, inspire us to never stop embracing our uniqueness. Mentorship and Career Growth We’re continuously raising our performance bar as we strive to become Earth’s Best Employer. That’s why you’ll find endless knowledge-sharing, mentorship and other career-advancing resources here to help you develop into a better-rounded professional. Key job responsibilities Our scientists and engineers collaborate across diverse teams and projects to offer state of the art, cost effective solutions for the control of Amazon quantum processor systems. You’ll bring a passion for innovation, collaboration, and mentoring to: Solve layered technical problems, often ones not encountered before, across our hardware stack. Develop requirements with key system stakeholders, including quantum device, test and measurement, and cryogenic hardware teams. Design, implement, test, deploy, and maintain innovative solutions that meet both strict performance and cost metrics. Research enabling control system technologies necessary for Amazon to produce commercially viable quantum computers.
Applied Scientist, Artificial General Intelligence, AGI Data Services
US, MA, Boston
The Artificial General Intelligence (AGI) team is seeking a dedicated, skilled, and innovative Applied Scientist with a robust background in machine learning, statistics, quality assurance, auditing methodologies, and automated evaluation systems to ensure the highest standards of data quality, to build industry-leading technology with Large Language Models (LLMs) and multimodal systems. Key job responsibilities As part of the AGI team, an Applied Scientist will collaborate closely with core scientist team developing Amazon Nova models. They will lead the development of comprehensive quality strategies and auditing frameworks that safeguard the integrity of data collection workflows. This includes designing auditing strategies with detailed SOPs, quality metrics, and sampling methodologies that help Nova improve performances on benchmarks. The Applied Scientist will perform expert-level manual audits, conduct meta-audits to evaluate auditor performance, and provide targeted coaching to uplift overall quality capabilities. A critical aspect of this role involves developing and maintaining LLM-as-a-Judge systems, including designing judge architectures, creating evaluation rubrics, and building machine learning models for automated quality assessment. The Applied Scientist will also set up the configuration of data collection workflows and communicate quality feedback to stakeholders. An Applied Scientist will also have a direct impact on enhancing customer experiences through high-quality training and evaluation data that powers state-of-the-art LLM products and services. A day in the life An Applied Scientist with the AGI team will support quality solution design, conduct root cause analysis on data quality issues, research new auditing methodologies, and find innovative ways of optimizing data quality while setting examples for the team on quality assurance best practices and standards. Besides theoretical analysis and quality framework development, an Applied Scientist will also work closely with talented engineers, domain experts, and vendor teams to put quality strategies and automated judging systems into practice.
Member of Technical Staff, Engineering Manager , Amazon AGI Lab
US, CA, San Francisco
Amazon launched the AGI Lab to develop foundational capabilities for useful AI agents. We built Nova Act - a new AI model trained to perform actions within a web browser. The team builds AI/ML infrastructure that powers our production systems to run performantly at high scale. We’re also enabling practical AI to make our customers more productive, empowered, and fulfilled. In particular, our work combines large language models (LLMs) with reinforcement learning (RL) to solve reasoning, planning, and world modeling in both virtual and physical environments. Our lab is a small, talent-dense team with the resources and scale of Amazon. Each team in the lab has the autonomy to move fast and the long-term commitment to pursue high-risk, high-payoff research. We’re entering an exciting new era where agents can redefine what AI makes possible. We’d love for you to join our lab and build it from the ground up! Key job responsibilities This role will lead a team of SDEs building AI agents infrastructure from launch to scale. The role requires the ability to span across ML/AI system architecture and infrastructure. You will work closely with application developers and scientists to have a impact on the Agentic AI industry. We're looking for a Software Development Manager who is energized by building high performance systems, making an impact and thrives in fast-paced, collaborative environments. About the team Check out the Nova Act tools our team built on on nova.amazon.com/act
Data Scientist, AWS Quick Data
US, CA, Santa Clara
Amazon Quick Suite is an enterprise AI platform that transforms how organizations work with their data and knowledge. Combining generative AI-powered search, deep research capabilities, intelligent agents and automations, and comprehensive business intelligence, Quick Suite serves tens of thousands of users. Our platform processes thousands of queries monthly, helping teams make faster, data-driven decisions while maintaining enterprise-grade security and governance. From natural language interactions with complex datasets to automated workflows and custom AI agents, Quick Suite is redefining workplace productivity at unprecedented scale. We are seeking a Data Scientist II to join our Quick Data team, focusing on evaluation and benchmarking data development for Quick Suite features, with particular emphasis on Research and other generative AI capabilities. Our mission is to engineer high-quality datasets that are essential to the success of Amazon Quick Suite. From human evaluations and Responsible AI safeguards to Retrieval-Augmented Generation and beyond, our work ensures that Generative AI is enterprise-ready, safe, and effective for users at scale. As part of our diverse team—including data scientists, engineers, language engineers, linguists, and program managers—you will collaborate closely with science, engineering, and product teams. We are driven by customer obsession and a commitment to excellence. Key job responsibilities In this role, you will leverage data-centric AI principles to assess the impact of data on model performance and the broader machine learning pipeline. You will apply Generative AI techniques to evaluate how well our data represents human language and conduct experiments to measure downstream interactions. Specific responsibilities include: * Design and develop comprehensive evaluation and benchmarking datasets for Quick Suite AI-powered features * Leverage LLMs for synthetic data corpora generation; data evaluation and quality assessment using LLM-as-a-judge settings * Create ground truth datasets with high-quality question-answer pairs across diverse domains and use cases * Lead human annotation initiatives and model evaluation audits to ensure data quality and relevance * Develop and refine annotation guidelines and quality frameworks for evaluation tasks * Conduct statistical analysis to measure model performance, identify failure patterns, and guide improvement strategies * Collaborate with ML scientists and engineers to translate evaluation insights into actionable product improvements * Build scalable data pipelines and tools to support continuous evaluation and benchmarking efforts * Contribute to Responsible AI initiatives by developing safety and fairness evaluation datasets About the team Why AWS? Amazon Web Services (AWS) is the world’s most comprehensive and broadly adopted cloud platform. We pioneered cloud computing and never stopped innovating — that’s why customers from the most successful startups to Global 500 companies trust our robust suite of products and services to power their businesses. Inclusive Team Culture Here at AWS, it’s in our nature to learn and be curious. Our employee-led affinity groups foster a culture of inclusion that empower us to be proud of our differences. Ongoing events and learning experiences, including our Conversations on Race and Ethnicity (CORE) and AmazeCon conferences, inspire us to never stop embracing our uniqueness. Mentorship & Career Growth We’re continuously raising our performance bar as we strive to become Earth’s Best Employer. That’s why you’ll find endless knowledge-sharing, mentorship and other career-advancing resources here to help you develop into a better-rounded professional. Work/Life Balance We value work-life harmony. Achieving success at work should never come at the expense of sacrifices at home, which is why we strive for flexibility as part of our working culture. When we feel supported in the workplace and at home, there’s nothing we can’t achieve in the cloud. Hybrid Work We value innovation and recognize this sometimes requires uninterrupted time to focus on a build. We also value in-person collaboration and time spent face-to-face. Our team affords employees options to work in the office every day or in a flexible, hybrid work model near one of our U.S. Amazon offices.
Applied Scientist, Artificial General Intelligence, AGI Data Services
US, MA, Boston
The Artificial General Intelligence (AGI) team is seeking a dedicated, skilled, and innovative Applied Scientist with a robust background in machine learning, statistics, quality assurance, auditing methodologies, and automated evaluation systems to ensure the highest standards of data quality, to build industry-leading technology with Large Language Models (LLMs) and multimodal systems. Key job responsibilities As part of the AGI team, an Applied Scientist will collaborate closely with core scientist team developing Amazon Nova models. They will lead the development of comprehensive quality strategies and auditing frameworks that safeguard the integrity of data collection workflows. This includes designing auditing strategies with detailed SOPs, quality metrics, and sampling methodologies that help Nova improve performances on benchmarks. The Applied Scientist will perform expert-level manual audits, conduct meta-audits to evaluate auditor performance, and provide targeted coaching to uplift overall quality capabilities. A critical aspect of this role involves developing and maintaining LLM-as-a-Judge systems, including designing judge architectures, creating evaluation rubrics, and building machine learning models for automated quality assessment. The Applied Scientist will also set up the configuration of data collection workflows and communicate quality feedback to stakeholders. An Applied Scientist will also have a direct impact on enhancing customer experiences through high-quality training and evaluation data that powers state-of-the-art LLM products and services. A day in the life An Applied Scientist with the AGI team will support quality solution design, conduct root cause analysis on data quality issues, research new auditing methodologies, and find innovative ways of optimizing data quality while setting examples for the team on quality assurance best practices and standards. Besides theoretical analysis and quality framework development, an Applied Scientist will also work closely with talented engineers, domain experts, and vendor teams to put quality strategies and automated judging systems into practice.
Applied Scientist, Artificial General Intelligence, AGI Data Services
US, MA, Boston
The Artificial General Intelligence (AGI) team is seeking a dedicated, skilled, and innovative Applied Scientist with a robust background in machine learning, statistics, quality assurance, auditing methodologies, and automated evaluation systems to ensure the highest standards of data quality, to build industry-leading technology with Large Language Models (LLMs) and multimodal systems. Key job responsibilities As part of the AGI team, an Applied Scientist will collaborate closely with core scientist team developing Amazon Nova models. They will lead the development of comprehensive quality strategies and auditing frameworks that safeguard the integrity of data collection workflows. This includes designing auditing strategies with detailed SOPs, quality metrics, and sampling methodologies that help Nova improve performances on benchmarks. The Applied Scientist will perform expert-level manual audits, conduct meta-audits to evaluate auditor performance, and provide targeted coaching to uplift overall quality capabilities. A critical aspect of this role involves developing and maintaining LLM-as-a-Judge systems, including designing judge architectures, creating evaluation rubrics, and building machine learning models for automated quality assessment. The Applied Scientist will also set up the configuration of data collection workflows and communicate quality feedback to stakeholders. An Applied Scientist will also have a direct impact on enhancing customer experiences through high-quality training and evaluation data that powers state-of-the-art LLM products and services. A day in the life An Applied Scientist with the AGI team will support quality solution design, conduct root cause analysis on data quality issues, research new auditing methodologies, and find innovative ways of optimizing data quality while setting examples for the team on quality assurance best practices and standards. Besides theoretical analysis and quality framework development, an Applied Scientist will also work closely with talented engineers, domain experts, and vendor teams to put quality strategies and automated judging systems into practice.
Applied Scientist, Artificial General Intelligence, AGI Data Services
US, MA, Boston
The Artificial General Intelligence (AGI) team is seeking a dedicated, skilled, and innovative Applied Scientist with a robust background in machine learning, statistics, quality assurance, auditing methodologies, and automated evaluation systems to ensure the highest standards of data quality, to build industry-leading technology with Large Language Models (LLMs) and multimodal systems. Key job responsibilities As part of the AGI team, an Applied Scientist will collaborate closely with core scientist team developing Amazon Nova models. They will lead the development of comprehensive quality strategies and auditing frameworks that safeguard the integrity of data collection workflows. This includes designing auditing strategies with detailed SOPs, quality metrics, and sampling methodologies that help Nova improve performances on benchmarks. The Applied Scientist will perform expert-level manual audits, conduct meta-audits to evaluate auditor performance, and provide targeted coaching to uplift overall quality capabilities. A critical aspect of this role involves developing and maintaining LLM-as-a-Judge systems, including designing judge architectures, creating evaluation rubrics, and building machine learning models for automated quality assessment. The Applied Scientist will also set up the configuration of data collection workflows and communicate quality feedback to stakeholders. An Applied Scientist will also have a direct impact on enhancing customer experiences through high-quality training and evaluation data that powers state-of-the-art LLM products and services. A day in the life An Applied Scientist with the AGI team will support quality solution design, conduct root cause analysis on data quality issues, research new auditing methodologies, and find innovative ways of optimizing data quality while setting examples for the team on quality assurance best practices and standards. Besides theoretical analysis and quality framework development, an Applied Scientist will also work closely with talented engineers, domain experts, and vendor teams to put quality strategies and automated judging systems into practice.
See more jobs
View from space of a connected network around planet Earth representing the Internet of Things.
Get more from Amazon Science
Subscribe to our monthly newsletter

Amazon.com | Conditions of Use | Privacy | © 1996-2026 Amazon.com, Inc. or its affiliates

Social