
Diagnoses and mitigates GKE GPU and TPU node disruptions from host maintenance events
- Open source
- 24.4K GitHub stars
- Cloud and infrastructure

Diagnoses and mitigates GKE GPU and TPU node disruptions from host maintenance events

Diagnoses GKE TPU and GPU nodes stuck NotReady after kernel panics or hardware lockups

Guides agents through diagnosing GKE TPU multi-slice training hangs using MXLA analyzer

Diagnoses GKE TPU SliceCreationFailed errors caused by overlapping partitions or nodes

Diagnoses GKE TPU training throughput drops using ML Diagnostics and Cloud Monitoring

Diagnoses GKE TPU node pool stockouts caused by unhealthy pools holding reservation slots

Diagnoses GKE TPU slice creation stalls caused by unstable Compute Engine MIGs

Diagnoses GKE TPU VMs stuck in Compute Engine REPAIRING state and routes recovery

Sets up Backup for GKE backup plans, restores and disaster recovery for clusters

Guides an agent through GKE cluster creation, kubectl credentials and Autopilot vs

Guides an agent through running batch, HPC and job-queue workloads on GKE

Diagnoses GKE Pods stuck Pending from Compute Engine capacity or quota errors

Guides creating and auditing GKE clusters from golden-path templates for Autopilot

Answers GKE cost questions by querying billing exports and live cluster usage

Guides an agent through lowering GKE cluster and workload costs with rightsizing and Spot

Finds the right GKE golden base image for building custom node images

Applies GKE production defaults, readiness checklists and config audits for clusters

Deploys and tunes LLM and AI inference servers on GKE with GPUs and TPUs

Generates hardened, GKE-ready Kubernetes YAML manifests for Autopilot and Standard

Plans and sets up namespace isolation, quotas, RBAC and cost tracking for GKE teams

Guides an agent through GKE cluster networking: private clusters, VPC-native setup

Guides an agent through diagnosing GKE nodes stuck in NotReady status

Guides an agent through GKE monitoring, logging, and Managed Prometheus setup

Guides an agent through hardening GKE cluster-level security settings and IAM roles

Orchestrates GKE production readiness reviews across scaling, security, reliability

Guides agents to configure GKE workload reliability with PDBs, health probes and zone

Guides GKE edge networking: Gateway API, Ingress, Cloud Armor, NEGs, PSC and LB

Finds GKE StatefulSets with PDBs but no readiness probes and proposes fixes

Configures GKE persistent disks, Filestore shares and StorageClasses for pods

Diagnoses GKE persistent storage failures like attach, mount, pressure and expansion
See every kind of AI for cloud and infrastructure: tools, agents, MCP servers and skills.Ranked: the best AI for cloud and infrastructure.