AI Infrastructure news and analysis on Web Pulse: 34839 curated articles tracking the latest AI Infrastructure developments, tools, and research, updated continuously from vetted sources.
The official Python MCP SDK's 2.0.0 release, published July 28, 2026, removed and renamed public APIs, breaking wrapper libraries that depended on `mcp` without an upper version bound. A developer reproduced the failures…
OpenGEMM, an open-source project from developer aramesh10, released CUDA GEMM kernels for NVIDIA's B200 GPU (sm_100a), installable via pip and requiring PyTorch 2.8+ and CUDA 12.9+. The library emits standalone CUDA kern…
AMD and Intel's joint AI Compute Extensions specification, ACEv1, was merged into the GCC 17 development branch on September 2, 2026, delivering up to 16 times the compute density of equivalent AVX10 operations. The spec…
Amazon Bedrock AgentCore now supports the Model Context Protocol (MCP) Apps extension, letting developers render interactive HTML widgets inside AI hosts such as Claude and ChatGPT, with the AgentCore runtime handling se…
A developer published a Python and SQLite lab demonstrating how agent tool retries can create duplicate jobs when a write commits but the acknowledgment never reaches the caller. The lab shows that reusing the same opera…
XAI's GrokBot, which entered beta on August 11, 2026, and Meta's Muse, launched September 8, 2026, have both shipped persistent cloud virtual machines for AI agents, giving each agent a dedicated browser, filesystem, and…
OpenAI published a post detailing how it rapidly scaled online storage to serve over 1 billion ChatGPT users, according to the company's own engineering write-up. The post, the first part of a series, drew 1 point and 0 …
ZLUDA is a drop-in replacement for CUDA that allows unmodified CUDA applications to run on non-NVIDIA GPUs with near-native performance. The project targets users who want to execute existing CUDA workloads without code …
Beam released Beta9, an open-source runtime for serverless AI workloads that it positions as an alternative to Modal, offering container cold starts in under a second and fan-out to hundreds of containers. The tool provi…
A developer outlined a layered architecture for AI agent authentication, arguing that no single protocol suffices and that discovery, workload identity, request authentication, and delegated authorization each address di…
Thieves hijacked AI hardware shipments in California, according to a report whose body contained only subscription and membership promotional text with no further details. No specific figures, dates, companies, or named …
Apex Grid, a regtech developer building tools for Nigerian microfinance banks, has detailed its hybrid architecture that pairs vector databases with a versioned "data substrate" to make AI-driven compliance decisions tra…
A developer documented that MuleSoft's Inference Connector fails with a connection-refused error when pointed at a local Ollama instance via http://localhost:11434/v1, and that switching the openAICompatibleURL to http:/…
NVIDIA Triton Inference Server is being integrated into modern ETL architectures to enable AI-driven data enrichment, dynamic batching, and multi-framework model serving across GPU and CPU environments. The approach, des…
A developer built and open-sourced MCPBouncer, a zero-dependency, local-first action firewall and live packet inspector for the Model Context Protocol (MCP). The tool acts as an inline, zero-trust proxy between AI client…
The MCP server io.github.wygogogo19/robotbase-mcp v0.3.0 was published to the official MCP registry on 2026-09-13, offering read-only BTC/XMR/ZEC/DOGE/LTC chain data to AI agents through 16 tools. The server is distribut…
A reproducible Windows CUDA compatibility stack built on ZLUDA v6-preview.69 and AMD HIP SDK 6.4 has been published on GitHub, validated only on the AMD Radeon RX 9060 XT (gfx1200). The setup passes cuda_check for nvcuda…
Version 1.6.2 of the Othernet — La Réunion MCP server was published to the official MCP registry on 2026-09-13 as a hosted remote over streamable-http, according to the mcpindex index. A semantic screen by an 8b model fo…
A developer traced two reported bugs in a retrieval-augmented generation system back to flawed assumptions in chunk ranking and agent-memory recency scoring, then uncovered a third bug in a newly built Kafka Connect inge…
The MCP server com.datasignalslab/datasignals-lab-mcp version 1.7.0 was published to the official MCP registry on 2026-09-13 as a hosted remote over streamable-http, offering 19 tools covering hiring signals, SEC, Congre…