Skip to main content
Speedtrain has a small vocabulary. These pages define each term precisely, in the order the concepts build on each other.

Content sources

Indexes and stage outputs — the data a task runs over.

Task configurations

The versioned document describing one repeatable content job.

AI specialists

Who the AI is, and which model it runs on.

Variants

The dimensions the output varies across.

Capabilities

Tools the model may call, and letting it see images.

Quality checks

The rules that score every output.

Quality gates

How scores aggregate into accept or reject.

Review sessions

One run of a configuration over an input.

Destinations

Where accepted content is delivered.

Workflows

Connecting a source to a task for continuous processing.

​
The one-sentence version

A task configuration describes a content job. A review session runs a version of it over a content source, generating one output per item per variant combination. Quality checks score each output, a quality gate turns those scores into a verdict, a person decides, and accepted content goes to a destination.