Amazon

The article shows how to deploy llama.cpp on Amazon SageMaker AI using the AWS Deep Learning Container to provide OpenAI‑compatible chat completions. It explains extending the runtime with the SemIf module, which adds a probabilistic decision layer that returns option probabilities over a user‑defined set of

Amazon announced a suite of September 2026 enhancements for Bedrock, Bedrock AgentCore, and Strands that expand model availability, accelerate serverless agent runtimes with integrated evaluation, and enable automated knowledge‑base syncing through native enterprise connectors. These updates aim to streamline the deployment of multi‑model, evaluation‑enabled agents while reducing operational overhead for enterprise data integration.

Postman demonstrates how to run an AI agent at scale on Amazon Bedrock by adopting a set of architectural patterns that tame tool sprawl, expose schema‑based reads, and treat context size as the primary bottleneck. The approach allows a single agent to serve 40 million developers by limiting the number of integrated tools, using OpenAPI‑style schemas to fetch data, and carefully managing Bedrock’s context window to keep latency low. The result is a production‑ready agent that can be deployed at petabyte‑scale workloads without sacrificing responsiveness.

Amazon Bedrock AgentCore payments introduces a managed, on‑demand billing model for AI agents, allowing them to pay for each inference request and enforce spending limits directly within the Bedrock infrastructure. Incarna’s agents now pay BlockRun on a per‑request basis for model inference, handling 402 requests in a single run and eliminating the months‑long effort that would otherwise be required to add payment support manually. This integration demonstrates how Bedrock’s payment API can be leveraged to streamline cost control and operational efficiency in agent‑driven workloads.

The article outlines a reference architecture that lets multiple teams share a single Amazon SageMaker HyperPod EKS cluster while maintaining strict isolation and fairness. By leveraging AWS IAM Identity Center for authentication, assigning each team its own SageMaker Domain and Kubernetes namespace, and enabling HyperPod Task Governance, the design guarantees that workloads are isolated, scheduled fairly, and billed accurately at the namespace level. This approach reduces cluster fragmentation and simplifies cost tracking for large multi‑tenant AI workloads.

Claude Haiku 5.5 has been released on Amazon Bedrock and the Claude Platform on AWS, positioning itself as the fastest and most efficient member of the Claude 5.5 family. It is specifically engineered for subagent workflows and high‑volume, cost‑sensitive applications, delivering performance gains while cutting costs by roughly 75 % compared to Claude Haiku 4.5 on most tasks. The model’s integration into Bedrock allows developers to invoke it via the standard Bedrock API, while the Claude Platform offers a managed interface for fine‑tuning and deployment. The announcement highlights a clear trade‑off: the model prioritizes speed and cost efficiency, potentially at the expense of some advanced reasoning capabilities found in larger Claude variants.

Amazon Bedrock’s new Knowledge Base feature, coupled with Amazon Quick, introduces real‑time, document‑level access control for enterprise RAG workflows. By querying the source system’s author metadata during each retrieval

Qlik deployed its Qlik Answers platform on Amazon Bedrock, enabling 40,000+ customers to receive grounded, sourced answers from both structured and unstructured enterprise data. The solution uses a layered, multi‑agent architecture that routes inference requests across regions and leverages Bedrock Guardrails to enforce data safety and compliance. The design balances global scale with trust, but requires careful configuration of cross‑region routing and guardrail policies to maintain latency targets.

AWS DevOps Agent can diagnose incidents but stays in observe‑and‑report mode; the article demonstrates how to chain its investigation summaries through AWS Lambda Durable Functions, Amazon EventBridge, and Amazon Bedrock to automatically generate and deploy pre‑validated remediation actions. The solution orchestrates a durable workflow that listens for DevOps Agent alerts, invokes Bedrock to synthesize fix scripts, and pushes the validated changes back into the infrastructure, effectively closing the loop from detection to correction. The approach trades off a modest increase in orchestration latency for fully automated, AI‑driven incident response that can be scaled across multiple services.

Cornerstone OnDemand’s Orion AI, a multi‑agent system built on Amazon Bedrock and Strands Agents, slashed database diagnosis time from 45 minutes to 10 minutes—a 78 % reduction—within six months. The architecture couples Bedrock’s foundation LLMs with Strands’ agent orchestration to shift database operations from reactive firefighting to proactive automation. The result is a lean, three‑person team that can diagnose issues in a fraction of the time, trading off increased agent complexity for significant operational efficiency.

The post demonstrates how to build a persistent, context-aware personal assistant by combining Amazon Bedrock’s AgentCore runtime with OpenClaw. By leveraging AgentCore’s memory component, each chat session is stored as durable, structured knowledge that can be queried later via metadata tags, turning otherwise disposable conversations into reusable context. The integration shows how to attach metadata to messages with OpenClaw and retrieve relevant context across sessions, enabling a more personalized assistant experience.

Amazon SageMaker HyperPod administration is now streamlined through SageMaker Unified Studio, enabling platform teams to enforce cluster governance, design infrastructure boundaries, and allocate shared capacity. The guide details how to configure HyperPod clusters, set up IAM policies for

Amazon has added native SageMaker Studio support for managing SageMaker HyperPod Spaces on EKS clusters, letting users create, configure, start, stop, and open spaces directly from the Studio UI. The new workflow eliminates the need for CLI commands, enabling instant launch of JupyterLab or the Code Editor with just a few clicks. This integration streamlines the development cycle, but it is limited to HyperPod-enabled EKS clusters and does not yet expose advanced cluster‑level tuning options.

The authors showcase a voice‑first travel concierge built on Amazon Bedrock AgentCore, using Nova Sonic for real‑time speech-to‑text and text-to-speech, and Bedrock Knowledge Bases to answer policy queries. The agent invokes backend airline services through MCP tool handlers, enabling seat changes, delay checks, and

GLM 5.3, a 753‑billion‑parameter mixture‑of‑experts model from Z.ai, has been released on Amazon Bedrock, targeting coding and long‑horizon agentic tasks. The model can be invoked through Bedrock’s OpenAI‑compatible API, and Bedrock’s prompt‑caching feature cuts both cost and latency for repeated prompts. An open‑source security test suite is provided to validate the model’s compliance before deployment. The release demonstrates how Bedrock’s managed service can host extremely large, expert‑based LLMs while offering familiar API semantics and cost‑optimization hooks.