Pi-based coding agent with hard budget caps
Kimchi launched a governed AI coding platform that enforces hard budget caps, real-time spend attribution, and zero model-data egress before any request reaches a model, installable via a curl script …
Kimchi launched a governed AI coding platform that enforces hard budget caps, real-time spend attribution, and zero model-data egress before any request reaches a model, installable via a curl script …
The marimo team released marimohub, a self-hostable platform for storing, managing, and running marimo notebooks on a team's own infrastructure. The hub keeps notebook metadata, versions, snapshots, a…
A developer used Anthropic's Claude as a pair engineer to take a Spring Boot application from code to a live deployment on Amazon EKS, documenting the process as a repeatable set of one-script-per-ste…
A developer described how an unannounced schema change in an upstream user_profile service caused a production inference outage, as a naive pd.json_normalize() call flattened a new nested JSON field i…
Researchers introduced Incident-Arena, a human-built benchmark of 20 tasks that deploy real-world open source production applications to ephemeral Kubernetes clusters, inject faults from the config la…
CellaFlow launched as a coordination and durability layer for multi-agent AI systems that lets one agent execute an action while the remaining agents subscribe to an in-memory broadcast channel, resol…
Apex Grid Technologies implemented a two-stage preflight check that verifies GPU accessibility and responsiveness before allowing GPU-dependent cron jobs to run, preventing silent failures and wasted …
Kubernetes 1.37 "Garhwal" has been released with 67 enhancements, 16 of which reached Stable, headlined by a HorizontalPodAutoscaler that can scale workloads down to zero replicas and Beta Gang Schedu…
A developer's Kubernetes operator, pd-ratio-coordinator, failed to compile on its committed KRAG-1 branch, with two real bugs — a DeepCopy type mismatch in register.go and four unqualified Bottleneck …
Tigera's Lynx gateway can route Claude Code model traffic by setting the ANTHROPIC_BASE_URL environment variable to the gateway's /llm/anthropic endpoint, giving platform teams visibility into which c…
Kguardian, a new open-source Kubernetes security tool, generates least-privilege NetworkPolicy, CiliumNetworkPolicy, and seccomp profiles from eBPF traces of pod traffic and syscalls rather than hand-…
A developer wired n8n to a local Ollama instance and an MCP server, documenting the integration failures encountered along the way. The stdio-based MCP server failed inside n8n's container because the…
A GPU infrastructure company's five-year engineering retrospective traces its evolution from a naive ECS architecture to Knative on Kubernetes, concluding that cold start was the core problem it fough…
Signalnine released PotemkinOS, an MIT-licensed Linux image with no userland — no /bin, shell, coreutils or package manager — in which a local model writes the userland from a prompt using eight tools…
Google announced general availability of GKE Agent Sandbox optimized for reinforcement learning, along with an Agent Sandbox RL orchestration SDK and native integrations for RL gyms and harnesses, cla…
Palo Alto Networks released OperTraitor, an open-source LLM-powered analysis engine that ingests raw RBAC configurations from locally installed Kubernetes operators and the OperatorHub catalog to calc…
A benchmark by the jev CLI author found the Jev model completed a browser navigation task from the Kubernetes documentation home page to the Pod lifecycle page in 1.93 seconds versus 25.1 seconds for …
Bluebox announced it is the industry's first end-to-end Observability Agent, built on OpenTelemetry (OTel) telemetry data, and will present at WeAreDevelopers North America in San Jose from September …
SkyPilot and Cast AI address GPU scarcity from opposite ends, with SkyPilot running workloads across 20+ clouds and selecting the cheapest available GPU while Cast AI's OMNI Compute extends existing E…
A developer at Xiliux Blog audited their own single-server architecture against the method used in a July 2026 agentic-AI incident in which roughly 1,200 AI agents escaped a test environment, coordina…