AWS
Amazon Web Services releases and Terraform AWS provider.
- AWS What's New aiawsengineerhealthcare ·
AWS HealthOmics Adds Support for Nextflow 26.04
AWS HealthOmics now supports Nextflow version 26.04, introducing features like record types and a strict syntax parser. These enhancements improve workflow readability, reduce errors, and potentially lower compute costs by enabling earlier detection of pipeline issues. The update benefits healthcare and life sciences customers using AWS HealthOmics for bioinformatics.
feature - AWS What's New aiawsengineer ·
Amazon Quick adds VPC connectivity for private MCP servers
Amazon Quick now supports connecting to privately hosted Model Context Protocol (MCP) servers via VPC. This allows organizations to securely integrate their proprietary applications and custom data sources with Quick's AI workflows without exposing servers to the public internet. The feature is now available in all regions where Quick is offered.
feature - AWS What's New aiawsgaengineer ·
Amazon Bedrock AgentCore Identity supports BYO secrets with AWS Secrets Manager
Amazon Bedrock AgentCore Identity now allows referencing existing AWS Secrets Manager secrets, giving customers full control over secret creation, governance, and compliance. Previously, secrets were service-managed, limiting customization like CMK encryption and tagging. This enhancement supports organizations with strict governance requirements by enabling them to manage secrets according to their policies.
feature - AWS What's New aiawsengineer ·
Amazon SageMaker HyperPod adds EFA-only network interfaces
Amazon SageMaker HyperPod now supports EFA-only network interfaces, allowing dedicated Elastic Fabric Adapter (EFA) devices without traditional Elastic Network Adapters (ENA). This change enables larger AI/ML clusters by preventing IP address exhaustion in VPCs, which is crucial for large-scale distributed training workloads.
feature - AWS What's New aiawsengineer ·
SageMaker HyperPod adds troubleshooting for AI coding assistants
Amazon SageMaker HyperPod now includes troubleshooting skills that integrate with AI coding assistants like Claude Code, Cursor, and Kiro. These skills automate cluster diagnostics for large-scale AI/ML infrastructure, reducing the complexity and time required for troubleshooting distributed training and inference issues. This feature benefits AI/ML engineers and architects managing complex distributed clusters by providing faster resolution times and actionable recommendations.
feature - AWS What's New aiobservabilityawsengineer ·
Amazon Bedrock adds CloudWatch metrics for OpenAI/Anthropic APIs
Amazon Bedrock now provides Amazon CloudWatch metrics for its bedrock-mantle endpoint, which supports OpenAI and Anthropic APIs. This enables customers to monitor inference traffic, set alarms, and plan capacity for their generative AI applications, similar to how they monitor other AWS services. The metrics are published under the AWS/BedrockMantle namespace and include counts, token totals, and error rates, with granularity down to model and project levels.
feature - AWS What's New aiawsgaengineer ·
OpenAI's GPT-5.5, GPT-5.4, and Codex available on Amazon Bedrock
Amazon Bedrock now supports GPT-5.5 and GPT-5.4 for production workloads and Codex for AI-powered software development. This integration offers OpenAI's most capable models with AWS's security and operational controls, while Codex is accessible via multiple integrations and can run inference through Bedrock. The models are priced at OpenAI's first-party rates and count towards AWS commitments, with regional availability details provided.
feature - OpenAI News aiawsgaengineer ·
OpenAI models available on AWS
OpenAI's frontier models and Codex are now generally available on AWS. This integration allows enterprises to build with OpenAI using existing AWS environments, controls, and procurement workflows, enabling faster deployment from evaluation to production.
feature - AWS What's New aiinfraawsengineer ·
AWS Resilience Hub Next Generation Achieves General Availability
AWS Resilience Hub's next generation is now generally available, enhancing resilience assessment for critical workloads. This update introduces a new application model, dependency discovery, generative AI for failure mode analysis, and organization-wide reporting, benefiting platform engineering and SRE teams. The improvements aim to provide deeper insights and more actionable recommendations for workload resilience.
feature announcement - AWS What's New aiawsengineer ·
Amazon Connect adds generative AI post-contact summaries in 8 new languages
Amazon Connect now generates AI-powered post-contact summaries in eight new languages, including Portuguese, French, Italian, German, Spanish, Chinese, Japanese, and Korean. This expansion also includes non-US English variations, allowing for more localized and efficient agent wrap-up and manager review across global support operations. The feature is now available in all AWS Regions where post-contact summaries are supported.
feature - AWS What's New aiawsgaengineer ·
Claude Opus 4.8 available on AWS
AWS now offers Claude Opus 4.8, Anthropic's most capable generally available model, bringing significant improvements in coding, knowledge work, and autonomous tasks. This model offers enhanced reasoning, consistency, and error recovery, making it suitable for production AI applications. Developers and enterprises can access Claude Opus 4.8 via Amazon Bedrock for AWS-integrated data security or the Claude Platform on AWS for a native Anthropic experience with AWS billing and authentication.
feature announcement - AWS What's New aiawsgaengineer ·
Amazon OpenSearch Serverless Next Generation Generally Available
AWS announced the general availability of the next generation of Amazon OpenSearch Serverless, featuring auto-scaling up to 20x faster and resource provisioning in seconds. This release decouples compute and storage, offers scale-to-zero pricing for cost savings, and simplifies network connectivity with new endpoints, making it ideal for agentic workflows and AI development platforms.
feature patch - AWS What's New aiawsengineer ·
Amazon Connect UI Builder integrates AI Assistant
Amazon Connect's AI assistant is now embedded in the UI builder, allowing managers to generate and modify contact center views using natural language prompts. This feature aims to significantly reduce the time and expertise required for building user interfaces, potentially by up to 70%. The assistant handles tasks from creating forms to configuring layouts and styling, simplifying the UI development process for managers.
feature - AWS What's New aiawsgaengineer ·
AWS SageMaker Notebooks Gain NVIDIA H100 GPU Instances in Tokyo
Amazon EC2 P5.48xl instances, featuring NVIDIA H100 GPUs, are now generally available on SageMaker notebook instances in the Asia Pacific (Tokyo) region. These instances offer up to 4x performance improvement and a 40% cost reduction for training ML models, benefiting users working with LLMs and generative AI applications.
feature - AWS What's New aiawsgaengineer ·
AWS SageMaker Notebooks Add NVIDIA Blackwell GPU Instances
Amazon SageMaker notebook instances now support EC2 P6-B200 instances powered by NVIDIA Blackwell GPUs in US East (N. Virginia). These instances offer up to 2x performance improvement for AI training, enabling efficient development and fine-tuning of large foundation models, including LLMs, for generative AI applications.
feature - AWS What's New aiawspreviewengineer ·
Amazon Bedrock adds Service Quotas support for bedrock-mantle endpoint
Amazon Bedrock now supports AWS Service Quotas for its bedrock-mantle endpoint, providing customers with visibility into inference quotas for OpenAI and Anthropic APIs. This allows for consistent tracking of limits across AWS services and helps customers proactively plan for production workloads.
feature - AWS What's New aiawsengineer ·
AWS Elemental Inference adds AI-powered live captioning
AWS Elemental Inference now supports Smart Subtitles, an AI feature for automatic live caption generation. This enables broadcasters to add real-time, low-latency TTML-formatted subtitles in multiple languages, improving accessibility without manual workflows. The feature integrates with AWS Elemental MediaLive and allows for custom dictionary creation to enhance transcription accuracy for specialized content.
feature - AWS What's New aiawsgadata-scientist ·
SageMaker Notebook Instances Gain P5.4xl GPU Instance Support
Amazon SageMaker notebook instances now support EC2 P5.4xl instances, featuring NVIDIA H100 GPUs. This offers up to 4x performance improvement and 40% cost reduction for training large language and generative AI models, benefiting ML engineers and data scientists working on complex AI applications.
feature - AWS What's New aiawsgaengineer ·
SageMaker Notebook Instances add P5en.48xl instance types with H200 GPUs
Amazon SageMaker notebook instances now support EC2 P5en.48xl instances, which feature H200 GPUs offering increased memory and bandwidth over H100 GPUs. This enhancement significantly boosts AI training and inference performance, especially for distributed workloads, and is available in select AWS regions. The new instances are suitable for developers and architects working with deep learning, generative AI, and HPC applications.
feature - AWS What's New aiawsgaengineer ·
Amazon Connect uses generative AI for automated interaction evaluation
Amazon Connect now offers generative AI capabilities allowing managers to automatically evaluate self-service interactions based on custom natural language criteria. This feature provides aggregated insights and detailed reasoning to help improve AI agent performance and customer experience. Managers can use these evaluations alongside recordings and transcripts to identify areas for improvement.
feature