Train, evaluate, deploy

A model's path to production as four bubbles on a liquid bar: data, training and evaluation fill one after another, the eval score counts to 92.4, and Deploy lights up for you to ship it.

Run stepsNo. 0622 variationsFree with an account

Playing on its own
Variation
The idea
A model's path to production
Autoplay
Liquid runs along the bar and fills each step bubble with a moving wave: Data, then Train with epochs and a falling loss curve, then Evaluate as the score counts to 92.4 and passes the gate; the Deploy bubble lights, the cursor deploys, then opens the eval step.
When you click
Click a step to open its panel; Deploy rolls the model out once the eval passes, then Run again restarts. Keyboard: Tab to the steps, Left and Right move between them, Enter on Deploy.
Look
stepper, liquid fill, sans type
Variations
Teal on ice, Signal
Files
index.html, style.css, script.js and prompt.md. No framework, no CDN, works offline.

The prompt and the files

Paste the prompt into Claude, ChatGPT, Lovable or any coding agent to rebuild it in your own product, or take the three files as they are.

Loading the files

More run steps

See all