As a Principal Engineer, Gaurav thrives on the fast-paced innovation and massive scale of Amazon's operations. He finds motivation in the challenges of operating at scale, creating opportunities for machine learning models to learn, and improving sampling techniques. "The new challenges I've been getting, I've never been bored," says Gaurav, highlighting the dynamic nature of his role. We celebrate the dedication and expertise of our engineers like Gaurav who drive innovation on behalf of Amazon's customers. To learn more about open roles and explore the Principal Engineering Community at Amazon, visit https://lnkd.in/gr6iB2yx. #AmazonPEs #InsideAmazon
Amazon Science
Research Services
Seattle, Washington 372,531 followers
The latest news and research from Amazon’s science community. #AmazonScience
About us
Amazon Science gives you insight into the company’s approach to customer-obsessed scientific innovation. Amazon fundamentally believes that scientific innovation is essential to being the most customer-centric company in the world. It’s the company’s ability to have an impact at scale that allows us to attract some of the brightest minds in artificial intelligence and related fields. Our scientists continue to publish, teach, and engage with the academic community, in addition to utilizing our working backwards method to enrich the way we live and work. Follow us on LinkedIn and visit our website to get a deep dive on innovation at Amazon, and explore the many ways you can engage with our scientific community. #AmazonScience
- Website
-
https://www.amazon.science
External link for Amazon Science
- Industry
- Research Services
- Company size
- 10,001+ employees
- Headquarters
- Seattle, Washington
- Founded
- 2020
- Specialties
- Artificial Intelligence, Machine Learning, Computer Vision, Cloud, Economics, Sustainability, AI, ML, Conversational AI, Natural Language Processing, NLP, Robotics, Security, Privacy, Information, Knowledge Management, Operations, Scientific Research, Search, Amazon, and Alexa
Updates
-
Amazon Science reposted this
Since launching Amazon Nova foundation models two months ago, we've seen incredible momentum and impact. I'm particularly impressed by what AppFolio has achieved.  AppFolio is a leading provider of software and services to the real estate industry and the creator of AppFolio Realm-X, an embedded generative AI tool that provides intelligent, real-time assistance for customers. Since integrating Amazon Nova Pro, the AppFolio team has found it to be a cost-effective solution for achieving superior performance.  Specifically: - AppFolio has seen an estimated 20% increase in adoption of Realm-X - Realm-X customers have reported saving over 10 hours of time per week on tasks on their to-do lists  Excited to see customers like AppFolio improving their experiences by building on the capabilities, speed and cost-effectiveness of Amazon Nova. Check out more about Amazon Nova here: https://amzn.to/4hJBFK6
-
A useful application of LLMs is to convert unstructured data into structured data. A pair of recent Amazon papers shows that a novel training procedure and decoding mechanism enable faster, lightweight LLMs to outperform larger LLMs on this task: https://amzn.to/4gvkCul #LargeLanguageModels #LLMs
-
Congratulations to Amazon Scholar Aaditya Ramdas on receiving two prestigious honors: the Presidential Early Career Award for Scientists and Engineers (PECASE), the highest honor bestowed by The White House on early-career scientists, and the Kavli Fellowship from the National Academy of Sciences, recognizing exceptional research achievements by scientists under 45. More on the awards: PECASE: https://amzn.to/3PLQAr3 Kavli Fellows: https://amzn.to/3DWYgnL
President Biden Honors Nearly 400 Federally Funded Early-Career Scientists
whitehouse.gov
-
“QKD doesn’t solve the authentication problem†is a misnomer, say AWS researchers: key exchange protocols and authentication mechanisms are primitives for solving distinct problems and must be integrated in a secure communication system.
Quantum key distribution and authentication: Separating facts from myths
amazon.science
-
At ASPLOS/EuroSys 2025, AWS is sponsoring a machine learning optimization challenge that invites teams to develop the fastest implementation of Llama 3.2 1B on Amazon ML hardware. Participants will leverage the Neuron Kernel Interface (NKI) to optimize model performance on AWS purpose-built AI chips, Trainium and Inferentia. Teams can win up to $25,000 for open-sourcing their solutions. Learn more:
Optimized NKI Implementation of Llama 3.2 1B (Inference)
github.com
-
Our most-viewed papers of 2024 span cloud databases, anomaly detection on graphs, recession prediction, and Amazon's new Nova foundation models. Explore the most popular publications authored by Amazon scientists and collaborators from this year: https://amzn.to/3VTj1qG #AmazonScience
-
As 2024 comes to a close, we've rounded up our 10 most-viewed blog posts of the year. Explore our breakthrough research in large language models, cryptography, automated reasoning, and more: https://amzn.to/3Pa95Fl #AmazonScience
-
Chronos-Bolt is our newest open-source foundation model for time series forecasting, which has been integrated into AutoGluon-TimeSeries. It delivers accurate zero-shot forecasts up to 250 times faster than the original Chronos models. Chronos-Bolt is based on the T5 encoder-decoder architecture and has been trained on nearly 100 billion time series observations.
Fast and accurate zero-shot forecasting with Chronos-Bolt and AutoGluon | Amazon Web Services
aws.amazon.com