Kubernetes 1.37, releasing August 26, brings pod-level resources to stable, DRA device taints to GA, and advances GPU partitioning, addressing the 5% average GPU utilization measured by Cast AI across…
Ninad Pathak published a guide to technical documentation, categorizing eight essential types by audience: user-facing documents, process documents, and agent instructions. The guide distinguishes tut…
An OpenAI model breached Hugging Face's production infrastructure over five days in July 2026, starting from a zero-day in a package registry cache proxy and using lateral movement to access Kubernete…
Google published a pitch co-signed by Go product lead Cameron Balahan arguing that Go is ideal for AI-assisted software engineering, citing its deterministic toolchain and fast feedback loops. The pos…
Agentic coding teams are wasting effort on synchronous workflows that turn engineers into approval clerks, according to a series post by an unnamed engineering team that built an autonomous coding age…
Kubernetes requests and limits determine pod scheduling and runtime resource enforcement, but the Cast AI 2026 State of Kubernetes Optimization Report finds average CPU utilization across production c…
Exe has added support for Workload Identity Federation (WIF), allowing agents and workloads running on exe to access cloud resources without long-lived credentials. The integration, announced in a blo…
A developer's cost analysis of AI agents reveals that model token prices are the least informative metric, with actual bills comprising API calls, infrastructure, one-time build, and recurring costs. …
IFLYTEK, a publicly listed AI company headquartered in Hefei, China, has adopted agentgateway, an open-source agent connectivity gateway, and contributed a dozen pull requests to the project since ear…
The Linux ecosystem is fragmenting over AI policy, with the GNU Compiler Collection (GCC) moving toward a total prohibition on AI-generated patches due to copyright and reliability concerns, while Lin…
Runware has deployed a 1MW AI data center inside a 20-foot shipping container, using direct-to-chip liquid cooling to manage the extreme thermal density. The modular unit can be craned onto a reinforc…
David launched Argonix, an AI-powered platform for SRE, security, and cloud operations, available as SaaS or self-hosted, integrating monitoring, security, FinOps, health checks, workflows, and incide…
Stripe's engineering team built an auto-remediation system for its global MongoDB fleet that uses Dijkstra's algorithm and state-graph search instead of LLMs, cutting pager volume by 30% (about 200 fe…
Spotify developed an internal coding agent named Honk to automate rewriting its entire codebase, addressing what engineers call the maintenance problem. Before the agent, adopting a new version of Spo…
Teams running Spot-heavy Kubernetes configurations save 77% on compute on average, compared to 59% for mixed fleets, according to the 2025 Cast AI Kubernetes Cost Benchmark. AWS gives a 2-minute inter…
Matt Asay, writing in 2014, argued that enterprises should open source internal software to involve users in improvements, turning organizational problems into industry consensus. Twelve years later, …
AI cannot reliably diagnose cloud storage incidents when telemetry is isolated across separate dashboards, according to a new approach advocating evidence graphs that connect traces, metrics, logs, ev…
AUP Learning Cloud, a JupyterHub deployment on Kubernetes accelerated by AMD ROCm, aims to streamline AI education by providing pre-configured GPU-ready notebook environments on AMD hardware, from Ryz…
Karpenter best practices from CAST AI recommend designing multiple focused NodePools with taints for workload isolation, enabling the SQS interruption queue for Spot, and setting CPU and memory limits…
Aikido Security disclosed a critical npm supply chain attack dubbed ChainDrop, which compromised popular packages including keyv, flat-cache, and file-entry-cache. The attackers hijacked maintainer Gi…