Serve AI models on GKE with the right hardware.
What it does
Deploys AI models on GKE with optimized inference manifests, accelerator selection, and autoscaling guidance.
When to use it
Use when deploying inference servers or LLMs on GKE, generating inference manifests, selecting accelerators, or configuring autoscaling.
How to use it
Give it an AI inference deployment request; it generates a manifest, deploys it to GKE, and monitors the workload.
Uses
GK
Latest review
What you need · 4
Requires a golden path GKE Autopilot cluster and sufficient GPU/TPU quota in the target region.
Requires an authenticated gcloud CLI for discovering inference profiles and generating manifests.
Access · 2
This skill
GKE
Execute
Starts inference workloads on GKE
Write
Applies manifests to GKE
Google Cloud
Read
Looks up supported inference configurations