Amazon

OPLOG rolled out a production BI system of three autonomous agents — built with the Strands Agents SDK and running on Amazon Bedrock AgentCore — that delivered a 35% shorter sales cycle, 91% improvement in CRM data

Amazon published a technical how‑to demonstrating an Agentic AI approach to optimize radiology worklists using Amazon Bedrock AgentCore and the Strands Agents SDK.

AWS published a solution blueprint showing how to combine Amazon Bedrock AgentCore, Strands Agents, and Amazon Quick to build multi‑agent systems that accept natural‑language requests to discover and modify dashboards,

A new how-to shows implementing Recursive Language Models (RLMs) with Amazon Bedrock AgentCore Code Interpreter and the Strands Agents SDK to analyze documents of virtually unlimited size.

A technical blog presents Amazon Bedrock AgentCore as a managed, serverless framework designed to help SaaS providers move multi‑tenant agentic applications from demos to production by addressing tenant isolation,

A new Amazon Bedrock reference architecture demonstrates how foundation models, coordinated through the Bedrock Converse API and combined with serverless AWS services, can automate parts of recruitment workflows such as

A new hands‑on walkthrough shows how to use Amazon Bedrock AgentCore Runtime with Model Context Protocol (MCP) support to link Amazon Quick to AWS services via a containerized AWS API MCP Server, translating

SageMaker AI now accepts OpenAI Chat Completions at an /openai/v1 path on real-time inference endpoints and supports streaming; developers can authenticate with SageMaker — generated bearer tokens (valid up to 12 hours)

AWS introduced four multimodal MLLM evaluators in the Strands Evals SDK-Overall Quality, Correctness, Faithfulness, and Instruction Following — that send an image plus query, response, and optional reference

Starting November 2025, Amazon SageMaker AI will support HTTP/2 bidirectional streaming to model containers, and vLLM’s Realtime API provides native WebSocket transcription that emits tokens incrementally for lower

A technical walkthrough and sample repository show how to add session — persistent conversational memory to Kiro CLI by implementing a custom Model Context Protocol (MCP) server that interfaces with Amazon Bedrock

Amazon published guidance outlining three PTC implementation patterns on Bedrock—a self-hosted Docker sandbox on ECS, a managed AgentCore Code Interpreter, and an Anthropic SDK-compatible proxy — so the model emits code

A technical guide shows how to prompt Amazon Nova 2 Lite for content moderation using structured (XML/JSON) and free-form prompts aligned to the MLCommons AILuminate Assessment Standard v1.

Aderant says adopting Amazon Quick gave its 38 — person Cloud Engineering team a unified AI search across six knowledge systems and automated documentation workflows, cutting search times by 90% and reducing article

A technical walkthrough shows how to implement four AWS Lambda — based custom code evaluators for a financial market‑intelligence agent, register them with Amazon Bedrock AgentCore, and run them in on‑demand

Advanced Prompt Optimization is a new console tool in Amazon Bedrock that rewrites and evaluates prompt templates using example inputs, ground — truth answers, and metric — driven feedback, and compares original

Assisted NLU uses large language models to improve Amazon Lex intent classification and slot resolution, offered in Primary and Fallback modes at no extra cost;

Amazon Bedrock AgentCore’s browser now accepts Chrome enterprise policy JSON and custom root CA certificates so administrators can restrict agent browsing, disable risky browser features, and permit validated HTTPS

A new technical post describes a reference architecture that pairs Amazon Nova 2 Sonic with Amazon Kinesis Video Streams WebRTC to reduce latency, handle multilingual conversational workflows, scale automatically,

AWS-backed AI Registry now integrates with Cisco AI Defense to register, discover and automatically scan Model Context Protocol (MCP) servers, Agent‑to‑Agent (A2A) agents and Agent Skills.

A technical walkthrough demonstrates a secure workflow that integrates Databricks Unity Catalog with Amazon SageMaker AI and Amazon EMR Serverless to fine‑tune models without losing centralized metadata, permissions,

Amazon Finance’s Finance Technology teams built an intelligent regulatory‑response system using Retrieval‑Augmented Generation with Bedrock Knowledge Bases, Amazon OpenSearch Serverless vector storage, and real‑time chat

The EU AI Act now requires organizations that fine‑tune LLMs to track floating‑point operations (FLOPs) to determine whether they must register as general‑purpose AI (GPAI) providers;

Vapi closed a $50 million Series B led by Peak XV Partners at an implied $500 million valuation after Amazon Ring picked Vapi over 40 rivals and now routes 100% of inbound calls through its platform;