CoreWeave
Official@coreweave · New York
Accelerating The World
Agent Skills by CoreWeave
Showing 14 vetted skills indexed across 1 GitHub repositories.
coreweaveconfigure-sunk-user-auth
Configure OpenLDAP and nsscache for SUNK user authentication across Slurm pods.
coreweaveconfigure-sunk-authentik-sssd
Configure SUNK user authentication with Authentik and SSSD.
coreweavesunk-core-dump
Generate a comprehensive diagnostic dump of SUNK cluster resources and state.
coreweaveconnect-skypilot-to-sunk
Integrate SkyPilot with SUNK clusters for GPU workload submission.
coreweaveupgrade-sunk
Automate SUNK deployment upgrades with Helm pre- and post-upgrade steps.
coreweavesetup-sunk-gpu-monitoring
Set up GPU and Slurm metrics collection on a SUNK cluster.
coreweavebootstrap-sunk-local-user
Create local UNIX users with SSH keys and register them with Slurmdbd on SUNK deployments.
coreweavedeploy-sunk-on-gke
Automate SUNK deployment on Google Kubernetes Engine with cluster and Slurm setup.
coreweavepatch-gke-system-tolerations
Apply tolerations to GKE system pods for the SUNK lock taint.
coreweaveadd-gpu-nodes-to-gke
Automate GPU node pool creation, driver setup, and Slurm configuration on GKE.
coreweavedeploy-sunk-on-eks
Automate SUNK deployment on Amazon EKS with cost-optimized defaults.
coreweavepatch-eks-system-tolerations
Patches system pods to tolerate the SUNK lock taint in EKS clusters.
coreweaveadd-gpu-nodes-to-eks
Automate GPU nodegroup addition to EKS SUNK deployments with quota checks and Slurm configuration.
coreweavedeploy-sunk-on-kubernetes
Automate SUNK deployment on Kubernetes clusters with Helm and storage configuration.
Frequently Asked Questions About CoreWeave
FAQPage SchemaWhat specific infrastructure tasks does CoreWeave enable for cluster management?▼
CoreWeave enables the deployment of Slurm-based computing environments on Kubernetes, including GPU node pool provisioning, identity management configuration via OpenLDAP or SSSD, and cluster-wide diagnostic state collection. It streamlines the integration of high-performance scheduling with existing cloud-native infrastructure.
Which engineering personas benefit from these cluster management capabilities?▼
These capabilities are designed for Site Reliability Engineers, HPC Infrastructure Architects, and Cloud Platform Engineers. These professionals manage large-scale GPU resources and require standardized methods to integrate Slurm scheduling with managed Kubernetes environments like GKE and EKS.
What are the primary prerequisites for deploying these cluster configurations?▼
Deployments require an existing Kubernetes cluster, such as GKE or EKS, and Helm for package management. Users must also have administrative access to configure node taints, tolerations, and identity provider settings to ensure proper Slurm integration and secure user authentication.