agent infrastructure
30 articles about agent infrastructure in AI news
Claude Managed Agents: The DIY Cost Formula Every Developer Needs
A real-world cost breakdown shows when to use Claude Managed Agents vs. running your own multi-agent infrastructure, with a clear formula to decide.
Meta Deploys Unified AI Agents to Manage Hyperscale Infrastructure
Meta's engineering team has built and deployed a system of unified AI agents to autonomously manage capacity and performance across its hyperscale infrastructure. This represents a significant shift from rule-based automation to AI-driven orchestration for one of the world's largest computing fleets.
Visa Launches Global AI Agent Shopping Infrastructure
Visa is launching a global infrastructure to enable AI agents to shop and transact autonomously. This move, alongside reports of a 25% conversion uplift from Frasers Group's AI assistant, signals the acceleration of 'agentic commerce'.
Lyzr's $250M Valuation Signals Enterprise Shift to Agentic AI Infrastructure
Lyzr AI, a startup building infrastructure for enterprise AI agents, has raised funding at a $250 million valuation, led by Accenture. The investment highlights growing corporate demand for autonomous AI systems that can perform complex tasks without constant human oversight.
OpenAI's WebSocket Breakthrough: The Infrastructure Shift Making AI Agents 40% Faster
OpenAI has launched WebSocket Mode for its Responses API, enabling persistent connections that reduce redundant data transmission in AI agent workflows. This architectural change cuts latency by up to 40% for complex tool-calling operations, marking a significant infrastructure evolution for production AI systems.
Agent Harnessing: The Infrastructure That Makes AI Agents Work
A detailed technical guide argues that the model is not the hard part of building AI agents. The six-component harness — context management, memory, tools, control flow, verification, and coordination — is what separates production-grade agents from those that fail silently.
MiniMax Launches MMX-CLI, First Infrastructure Built for AI Agents
MiniMax released MMX-CLI, a CLI built for AI agents, not humans. It provides agents with seven multimodal 'senses' and native integration with popular AI coding environments.
Google DeepMind's Intelligent Delegation Framework: The Missing Infrastructure for AI Agents
Google DeepMind has introduced a groundbreaking framework called Intelligent AI Delegation that enables AI agents to safely hand off tasks to other agents and humans. The system addresses critical issues of accountability, transparency, and reliability in multi-agent systems.
Alibaba's OpenSandbox: The Free Infrastructure Revolution for AI Agents
Alibaba has open-sourced OpenSandbox, a production-grade sandbox environment for AI agents that provides secure code execution, web browsing, and model training capabilities with unified APIs across multiple programming languages.
Meta Deploys Millions of Amazon Graviton CPUs for AI Agents
Meta will deploy tens of millions of AWS Graviton5 CPU cores for AI agent workloads, signaling that agentic inference favors CPUs over GPUs. The deal deepens Meta's $200B+ infrastructure push amid layoffs and cloud rivalry.
NVIDIA, Google Cloud Expand AI Partnership for Agentic & Physical AI
NVIDIA and Google Cloud announced an expanded partnership to advance agentic and physical AI, focusing on new infrastructure and software integrations. This builds on their existing collaboration to provide optimized AI training and inference platforms.
Nvidia: Cost Per Token Is the Only AI Infrastructure Metric That Matters
Nvidia asserts that total cost of ownership for AI infrastructure must be measured in cost per delivered token, not raw compute metrics. This shift is critical for scaling profitable agentic AI applications.
Agent Harness Debate: Anthropic vs. OpenAI vs. LangChain on Scaffolding
A central debate in agent engineering pits a 'thin harness' approach (Anthropic) against 'thick harness' designs (LangGraph). The infrastructure layer, not the model, is becoming the primary product differentiator.
Google Ads Details Its Data Infrastructure for AI-Powered Commerce
Google Ads has detailed the critical role of its underlying product data infrastructure in enabling 'agentic commerce'—where AI agents assist shoppers. This foundation is key to making search more natural and understanding shopper intent.
Anthropic's Claude Code Now Acts as Autonomous PR Agent, Fixing CI Failures & Review Comments in Background
Anthropic has transformed Claude Code into a persistent pull request agent that monitors GitHub PRs, reacts to CI failures and reviewer comments, and pushes fixes autonomously while developers are offline. The system runs on Anthropic-managed cloud infrastructure, enabling full repo operations without local compute.
Alibaba Targets $100B in AI and Cloud Revenue, Betting on 'Agentic AI' for Commerce
Alibaba announced a five-year goal to generate over $100B from its AI and cloud divisions, pivoting its strategy toward the 'agentic AI era' where autonomous agents can complete transactions. This comes amid a major reorganization and heavy investment in AI infrastructure.
Alibaba Open-Sources OpenSandbox: A gVisor/Firecracker-Based Execution Environment for AI Agent Security
Alibaba has open-sourced OpenSandbox, a general-purpose execution environment that isolates AI agents in secure runtimes like gVisor or Firecracker. The system includes a code interpreter, managed filesystem, and network controls to prevent agents from accessing host infrastructure.
OpenSWE Releases 45,000+ Executable Environments for Training SWE Agents, Achieves 66% on SWE-bench Verified
OpenSWE introduces a framework with over 45,000 executable environments for training software engineering agents, achieving 66% on SWE-bench Verified through quality filtering of multi-agent synthesized environments. The Docker infrastructure is open-sourced for full reproducibility.
Perplexity's OpenClaw Evolution: Building Secure AI Agents for Local Hardware
Perplexity AI has expanded its agent ecosystem to enable local hardware and cloud infrastructure to run AI agents securely, addressing vulnerabilities found in earlier OpenClaw implementations while maintaining open-source accessibility.
Claude Code Wipes 2.5 Years of Production Data: A Developer's Costly Lesson in AI Agent Supervision
A developer's routine server migration using Claude Code resulted in catastrophic data loss when the AI agent deleted all production infrastructure and backups. The incident highlights critical risks of unsupervised AI execution in production environments.
SkillNet: The First Systematic Framework for Creating and Evolving AI Agent Skills
Researchers have developed SkillNet, an open infrastructure that structures over 200,000 AI agent skills within a unified ontology. The system improves agent performance by 40% while reducing execution steps by 30%, treating skills as evolving assets rather than temporary solutions.
The Autonomous Company: How 14 AI Agents Are Running a Startup Without Human Intervention
Auto-Co introduces a fully autonomous AI company operating system where 14 specialized agents debate, decide, and ship software 24/7. Using Claude Code CLI and a simple bash loop, this open-source system has built its own infrastructure, documentation, and community presence across 12 self-improvement cycles.
GuardClaw: The Cryptographic Audit Trail That Could Make AI Agents Accountable
GuardClaw introduces cryptographically verifiable execution logs for AI agents, creating immutable records of autonomous actions. This open-source protocol could revolutionize accountability in AI systems performing financial trades, infrastructure changes, and critical operations.
Parallax: How Durable Streams Are Revolutionizing Multi-Agent AI Collaboration
Parallax introduces a novel approach to AI agent coordination using isolated, append-only logs. This CLI tool enables independent agent cohorts to collaborate without seeing each other's reasoning, with disagreement enforced at the infrastructure level rather than through prompting.
Nebius Makes $275M Bet on AI Agent Search with Tavily Acquisition
European cloud provider Nebius acquires AI search startup Tavily for $275 million, integrating agentic search capabilities into its AI cloud platform to challenge major players in the competitive AI infrastructure market.
Why Agentic AI Demands a New Architecture: Bain's Strategic Framework
Bain & Company argues that deploying agentic AI systems requires fundamentally new architectural thinking, moving beyond simple API calls to orchestrated workflows. This has significant implications for how luxury brands should plan their AI infrastructure investments.
Building an Agentic Enterprise Control Plane on Snowflake: A Technical Blueprint
Snowflake Intelligence and Cortex Code now enable a fully embedded agentic AI control plane. This article provides a tested, end-to-end blueprint for building a production-grade Streamlit dashboard that integrates five enterprise tables with six Cortex AI functions, all governed by existing data platform RBAC.
San Francisco Shop Runs Entirely by AI Agent
A shop in San Francisco is fully operated by an AI agent, replacing human cashiers and assistants. The concept points toward fully autonomous retail experiences, though details on the technology stack remain thin.
Cua Driver Open-Sourced: macOS Agent Control for Any App
Cua released Cua Driver as open-source, allowing agents like Claude Code and Codex to drive any macOS app through visual understanding and direct UI interaction.
Agentic storefronts: How AI agents are reshaping the shopping journey from
Major tech companies integrate AI agents into search and checkout; platforms like ChatGPT become primary shopping discovery channels. Agentic storefronts (e.g., Swap) guide shoppers end-to-end, getting smarter per session.