KubeCon + CloudNativeCon North America 2026
Minting your own tokens.
The server side of the token economy. Deploy and test LLMs with vLLM on Kubernetes using your own browser-based workspace.
Deploy vLLM on a single GPU
Inspect logs and scale to two GPUs with tensor parallelism
Explore separate prefill and decode
Your workshop workspace
Join the hands-on lab
Sign in to open your private marimo notebook. Use a GitHub account with a verified primary email.
After signing in, wait for your workspace to open, then follow the instructor's steps. Use the same GitHub account when you return.