$ articles --latest
AI articles
New posts from the AI blogs we follow, plus articles the community submits.
latest · AI Coding Tools · 23h ago · 1 min readManaged Deep Agents v0.10: Browser Use and Async SubagentsManaged Deep Agents v0.10 introduces browser use, letting agents work on websites that lack an API and hand control back so you can sign in. It also adds async subagents that run background tasks.via LangChain
Model Releases · 1d ago · 1 min readSpeed Up Image Generation with FLUX 3 Image Klein on NVIDIA Blackwell GPUsLearn how to reduce diffusion model inference latency when running FLUX Image Klein on NVIDIA Blackwell GPUs. The piece covers why sequential passes and high-resolution attention slow generation in design and image editing tools, and how to speed up the pipeline.via NVIDIA Technical BlogObservability · 1d ago · 8 min readBeyond code generation: A developer’s guide to agentic engineeringExplains how developers can use GitHub Copilot and related tools across planning, building, verifying, and observing software work. It covers scattered planning context, running tasks across devices and sessions, and shrinking pull requests so review is faster. Also covers automated quality, security, and dependency checks.via GitHub Blog AI and ML
More articles
Giving Back to the Open Source We Build OnLanceDB describes its contribution of ASOF join support to Apache DataFusion, which improves time-series joins. The post explains why the company contributes back to the open source projects it depends on.Observability · via LanceDB · 1d ago1 minAccelerating Reinforcement Learning Weight Refit with the Reshard CollectiveLearn how the Reshard collective speeds up weight refit in reinforcement learning post-training, where updated policy weights must be synced from training GPUs to separate inference GPUs in a disaggregated setup. The article covers why this transfer becomes a bottleneck and how the new approach reduces it.Inference · via NVIDIA Technical Blog · 1d ago1 minSageMaker Inference now supports Multi-Instance GPU to maximize GPU utilization for model endpointsExplains how Amazon SageMaker AI now uses NVIDIA Multi-Instance GPU to split one physical GPU into isolated partitions, letting you host more models per endpoint and choose right-sized GPU capacity. It covers the API fields AcceleratorPartitionLayouts and AcceleratorPartitionType, plus the trade-offs and measured results.Open Models · via AWS Machine Learning · 1d ago8 minTaxonomy of agentic primitives in Amazon QuickAmazon Quick's team describes a shared taxonomy for agentic work built on four questions: When, What, Who, and How. It covers triggers that start work, agents as persistent learning actors, and tools, skills, and workflows as capabilities. The post also explains how a single request like a daily Slack digest can be met several valid ways.AI Coding Tools · via AWS Machine Learning · 1d ago8 minServerless MCP server for AI agent data lake access on Amazon ECS FargateLearn how a serverless MCP server on Amazon ECS Fargate gives AI agents centralized, metadata-driven access to a data lake. The post covers the four-layer architecture, how new datasets are added with a JSON file instead of code, and how authentication and session state are handled.Agents · via AWS Machine Learning · 1d ago8 minGovern AI agent access with Okta Cross App Access and Amazon Bedrock AgentCore – Part 2Shows how to put AWS AgentCore Gateway in front of MCP servers so Okta Cross App Access grants are redeemed per target, with Cedar policies binding each tool to an agent and its delegated scope. Covers adding a second agent on Amazon EKS and reading authorization decisions through AgentCore Observability and CloudWatch.Observability · via AWS Machine Learning · 1d ago8 minGovern AI agent access with Okta Cross App Access and Amazon Bedrock AgentCore – Part 1Learn how to register an AI agent in Okta, connect it to a resource authorization server, and use Cross App Access with Amazon Bedrock AgentCore to govern which user-authorized actions the agent can take. The post also covers token validation, measured revocation timing, and the practices that worked in testing.Agents · via AWS Machine Learning · 1d ago9 minContext Engineering (Agent Memory) for AI Agents Starts with RetrievalExplains why AI agents struggle with clumsy searches, oversized results, and stale data when managing context. Covers the retrieval techniques LangChain's Deep Agents harness uses to load the right information into an agent's context window.Agents · via LangChain · 1d ago1 minBrowser Use in Managed Deep AgentsExplains how Managed Deep Agents can run a browser for an agent, using tools designed to keep browser use efficient in context. It also covers a handoff that lets you sign in yourself during a session.AI Coding Tools · via LangChain · 1d ago1 min