Blog_dumb

Amazon SageMaker Feature Store is a fully managed, purpose-built repository to store, share, and manage features for machine learning (ML) models. It now supports Apache Iceberg table format, streaming ingestion, scalable batch ingestion, and fine-grained access control through AWS Lake Formation. As organizations scale their machine learning platforms from experimentation to production, two operational challenges […]

Programmatic tool calling (PTC) is a paradigm shift in how large language models (LLMs) interact with external tools. In a traditional tool-calling workflow, each tool invocation requires a full round trip back to the model. The model calls a tool, receives the result, reasons about it, calls the next tool, and so on. For workflows

If you moderate user-generated content at scale, you need a system that catches policy violations accurately without over-flagging legitimate posts. A moderation system that misses harmful content puts you at risk, while one that flags too aggressively frustrates your audience. Every organization defines its own policies, so a single classifier rarely works for every use

This guest post is co-written by Angela Mapes and Adam Walker of Aderant. Aderant, a leading global provider of comprehensive business management software for the legal industry, transformed how its 38-person Cloud Engineering team supports Expert Sierra, its cloud-based legal practice management solution. By implementing Amazon Quick, Aderant has accelerated documentation processes and empowered its

Teams can integrate Atlassian Confluence Cloud with Amazon Quick to search and manage documentation without switching between multiple systems. When documentation lives in Confluence, but related data sits in other systems, teams waste time switching tools, re-searching for context, and manually gathering information. These interruptions slow decisions and create gaps between available knowledge and actionable

Special thanks to everyone who contributed to this launch: Stephanie Yuan, Lefan Zhang, Ritvika Pillai, Irene Wang, Carter Williams, T.J Ariyawansa, Gitika Jha, Shoaib Javed and the product leadership from Vivek Singh. Moving prototype agents to production requires measuring quality across multiple dimensions. Amazon Bedrock AgentCore Evaluations provides large language model (LLM)-as-a-Judge checks and extensible

Organizations that must restrict access to sensitive documents increasingly rely on AI-driven search and chat to help employees find answers across large repositories. Coarse-grained permissions that control access at the knowledge base level work well for many teams, but sensitive documents require more granular control to restrict specific documents or folders to authorized teams, individuals,

Improving bot accuracy in Amazon Lex starts with handling how customers communicate naturally. Your customers express the same request in dozens of different ways, combine multiple pieces of information in one sentence, and often speak ambiguously. The Assisted NLU (natural language understanding) feature in Amazon Lex helps you improve bot accuracy by handling these natural

This post was co-authored with Neevash Ramdial, Technical Marketing leader at Stream Building production-grade voice agents that feel natural and responsive is a complex engineering challenge. You must orchestrate speech-to-speech models, manage low-latency audio streaming, and handle connection lifecycle. You also need to deliver consistent experiences across web, mobile, and desktop applications. In this post,

Amazon Quick is an AI-powered unified intelligence service that brings together an organization’s data, structured data and unstructured enterprise content like documents, emails, and knowledge bases into a single service where anyone can explore, analyze, and take action. With over 40 application integrations, Quick bridges the last-mile gap between insights and action so users can