apex:eks-genai

Source

This page is generated from steering/commands/apex/eks-genai.md. Edit the source, not this page.

Run the APEX GenAI-on-EKS workflow — turn a team's GenAI/LLM workload goals into an opinionated, layer-by-layer Amazon EKS stack recommendation and build path.

<execution_context> @~/.claude/apex-steering/workflows/eks-genai.md </execution_context>

Follow the eks-genai workflow end-to-end. Detect the user's mode from their message and route into the matching row of the workflow's "How to Route Requests" table. Use the `eks-genai` skill for the 6-layer stack, the GPU-vs-Neuron decision, the JARK + vLLM + LiteLLM canonical reference, KV-cache tiering, cost levers, and the security baseline. Phases: 1) workload scoping, 2) compute and cluster selection (Layers 1-2), 3) serving/training, storage, observability, gateway (Layers 3-6), 4) security baseline, cost, and build path.