Posts

[MS] Using Azure Blob Storage as a durable filesystem for LangChain Deep Agents - devamazonaws.blogspot.com

Image
LangChain Deep Agents is an open-source agent harness with built-in capabilities for building LLM-powered agents and applications, including complex, multi-step workflows. A model can reason and generate responses, but it needs a harness to do useful work over time. The harness provides the tools and runtime that let the model retrieve the right context, take actions, and manage work across multiple steps. Deep Agents supplies that structure through planning, context management, a virtual filesystem, memory and skills, specialized subagents, and human approval points. Filesystems give agents a familiar way to organize and work with context. An agent can list and search files, read only what the current task needs, update artifacts, keep notes, and share work with subagents. This makes the filesystem useful as both a workspace and external memory for long-running tasks, a pattern LangChain highlights in its Deep Agents architecture . Deep Agents exposes this pattern through tools fo...
Post a Comment
Read more

[MS] Foundry hosted agent isolation with Microsoft Agent Framework - devamazonaws.blogspot.com

Foundry hosted agent isolation exposes two independent controls. User isolation identifies whose data may be used . At the same time, Foundry hosted session isolation identifies where code and files continue to live . For each control, Foundry can resolve the value or the application can provide it. Therefore, the available choices support direct callers, trusted middle tiers, application-managed sessions, and shared session pools. While Foundry hosted agents are generally available in Microsoft Foundry, the AgentServer SDKs and the Microsoft Agent Framework Foundry hosting packages for .NET and Python are still in pre-release versions. Foundry hosted agent isolation controls Control Represents Available choices User isolation The user represented by the current request The caller's Microsoft Entra identity, or a delegated identity supplied by a trusted middle tier Foundry hosted session isolation The VM-isolated sandbox, represented by agent_session_id A sessio...
Post a Comment
Read more

Amazon EVS now in scope for FedRAMP Class C - devamazonaws.blogspot.com

The Amazon Elastic VMware Service (EVS) service is now in scope for FedRAMP Class C (formerly Moderate baseline) in all United States Regions.  The   Federal Risk and Authorization Management Program (FedRAMP)  is a US government-wide program that delivers a standard approach to the security assessment, authorization, and continuous monitoring for cloud products and services. Amazon EVS lets you run the latest VMware Cloud Foundation (VCF) software directly within your Amazon Virtual Private Cloud (VPC) on EC2 bare-metal instances. You can set up a complete VCF environment in just a few hours, enabling rapid workload migration to AWS to help you eliminate aging infrastructure, reduce operational risks, and meet critical timelines for exiting your data center. To learn more, visit the FedRAMP  AWS services in scope page . To learn more about Amazon EVS, visit the  product detail page and user guide .  Post Updated on September...
Post a Comment
Read more

[MS] Comparing exception behavior of magic statics, std::call_once, and std::async - devamazonaws.blogspot.com

We've been comparing magic statics, std:: call_ once , and std:: async , but one thing we haven't considered is their behavior in the event of an exception. For magic statics, if an exception occurs during initialization of the static, then the static is considered not to have been initialized. The exception propagates, and the next time the function is called, the language will try to initialize the static again. For call_ once , if an exception occurs during execution of the lambda, then the call is considered not to have occurred. The exception propagates, so the next time you call call_ once with the same once_ flag , it will try to call it. But std:: async is different. If an exception occurs during execution of the invocable, then the exception is saved, and when you ask the future or shared future for the result, the exception is rethrown. It does not try to execute the invocable again. Let's summarize this in a table.   Before After success After exc...
Post a Comment
Read more

[MS] Knowledge cutoff is a poor proxy for model capability - devamazonaws.blogspot.com

Image
A model's knowledge cutoff marks the latest date its training data could cover. But its actual knowledge of a product might end years earlier or be absent entirely. I learned this while trying to answer what seemed like a straightforward question: how much does a model know about a product? If the model's knowledge ended around a particular date, I should have been able to translate that date to a product version. The model knows the product up to roughly version X. Anything introduced after that needs additional context. On paper, that would be useful when building agent extensions. You could establish what the model already knows and supply the missing documentation, saving the context window for the work. In practice, the results showed no consistent boundary. Looking for the line To test it, I used GPT-5.6 Luna with the GitHub Copilot SDK against two developer products: Dev Proxy and SharePoint Framework (SPFx). I fed the changelogs and release notes for both...
Post a Comment
Read more

[MS] Spec-Driven Development comes to Azure Cosmos DB: The First Database Extension for GitHub Spec Kit - devamazonaws.blogspot.com

Image
AI coding agents can write much of an application's code, but developers still need to review the decisions behind it. For a Cosmos DB application, that includes choosing partition keys, modeling access patterns, and configuring the client. Those decisions affect cost, performance, and reliability long after the code compiles. We've written before about how Azure Cosmos DB supports agents at query time , with tools and guidance for exploring and querying data. Today we're announcing the public preview of the Azure Cosmos DB extension for GitHub Spec Kit , the first database extension in its ecosystem. It brings Cosmos DB guidance into spec-driven development , helping you and your coding agent work through application design before implementation. What is spec-driven development? In spec-driven development (SDD), you work through requirements, design, and tasks before implementation. The agent produces a document at each stage that you can read, correct, and approve...
Post a Comment
Read more

Amazon EC2 X8i instances are now available in the South America (São Paulo) Region - devamazonaws.blogspot.com

Starting today, Amazon Elastic Compute Cloud (Amazon EC2) X8i instances are available in the South America (São Paulo) region. These instances are powered by custom Intel Xeon 6 processors available only on AWS. X8i instances are SAP-certified and deliver the highest performance and fastest memory bandwidth among comparable Intel processors in the cloud. They deliver up to 43% higher performance, 1.5x more memory capacity (up to 6TB), and 3.3x more memory bandwidth compared to previous generation X2i instances. X8i instances are designed for memory-intensive workloads like SAP HANA, large databases, data analytics, and Electronic Design Automation (EDA). Compared to X2i instances, X8i instances offer up to 50% higher SAPS performance, up to 47% faster PostgreSQL performance, 88% faster Memcached performance, and 46% faster AI inference performance. X8i instances come in 14 sizes, from large to 96xlarge, including two bare metal options. To get started, visit the AWS Management Cons...
Post a Comment
Read more