AI Infrastructure news and analysis on Web Pulse: 32738 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
Amazon SageMaker AI now supports bidirectional streaming for real-time inference, enabling developers to build voice applications that process audio and return transcriptions simultaneously over a single persistent conne…
A developer has published a step-by-step guide for running the open-source Hermes Agent AI assistant 24/7 on a virtual private server (VPS) for as little as $5 per month. The setup, which connects the agent to messaging …
NVIDIA released an open-source deep research blueprint called AI-Q that can be integrated as a portable skill into agent harnesses such as Claude Code, Codex, and LangChain Deep Agents. The skill allows these general-pur…
The PyTorch Docathon 2026, held from May 5 to May 19, resulted in over 150 merged pull requests after more than 260 registrants and 30 active participants contributed fixes, API documentation, and ExecuTorch documentatio…
The Agentic AI Foundation (AAIF) has formally adopted the MLOps Community as its user community, integrating the practitioner network into the Linux Foundation-backed organization. The move gives MLOps Community members …
Exa has raised a Series C round led by a16z to build a search engine designed specifically for AI agents, which the company says will soon search the web more than humans. The startup has spent five years developing infr…
Andreessen Horowitz (a16z) is leading Exa's Series C funding round to support the company's development of a search engine built specifically for artificial intelligence agents. Exa's technology aims to replace tradition…
A vulnerability in Kubeflow versions prior to 1.10 allows attackers with Contributor or kubeflow-edit roles to steal authorization tokens via overly permissive Istio permissions, enabling account takeover and data access…
Grok Build 0.1, a beta coding model trained for agentic coding, is now available on Vercel AI Gateway in early access. The model powers the Grok Build CLI app and can be accessed by setting the model to `xai/grok-build-0…
At Google I/O 2026, the company announced Gemini 3.5 Flash is now generally available, positioning it as its strongest model for agentic and coding workloads with 1M-token context and four thinking levels. Google also in…
SigNoz has released a guide for setting up observability and monitoring on Dify agent workflows using OpenTelemetry, enabling users to export traces and metrics to SigNoz for unified dashboards and alerting. The integrat…
Google's AI pricing has tripled annually, while OpenAI's flagship model rose after a period of apparent subsidy, and Anthropic held steady before cutting costs on its most powerful models. The three vendors now face tigh…
The software industry's reframing of Specification-Driven Development (SDD) as a new methodology for AI-augmented teams is merely a 30-year-old practice in a modern wrapper, according to a new analysis. The practice of w…
1Password and OpenAI partnered to integrate 1Password as a trusted access layer for OpenAI's Codex, enabling just-in-time, scoped credential management that keeps secrets outside the model's context window. The integrati…
Alibaba's Qwen3.6-27B dense model outperformed its own 397-billion-parameter MoE predecessor on SWE-bench Verified (77.2% vs. ~76.2%) while running on a single RTX 4090, challenging the prevailing MoE paradigm for models…
Vercel released an AI Gateway plugin for WordPress that provides access to hundreds of AI models from over 40 providers through a single API key. The plugin integrates with the new WordPress AI Client, which requires the…
NVIDIA launched verified agent skills to provide capability governance for autonomous AI agents, embedding transparency, provenance, security validation, and authenticity checks directly into the skill layer. The verifie…
OpenAI on Tuesday launched Guaranteed Capacity, a new offering that provides enterprise customers with long-term, guaranteed access to its compute resources for one- to three-year commitments. The program aims to give bu…
Prime Video engineers Billy Hickman and Lilia Abaibourova presented a solution at the MCP Dev Summit North America 2026 for AI agents overwhelmed by hundreds of tools in a single context window. Their approach reduces th…
At the Google I/O conference, NVIDIA and Google Cloud announced new resources for their joint developer community of over 100,000 members, including learning paths for JAX on NVIDIA GPUs and an NVIDIA Dynamo codelab for …