# TReqs

> TReqs is where machine-learning training work gets planned, reviewed, and run. A training request pins code at a commit, a workflow file, and a compute target; it is reviewed like a pull request, then queued as a training run that publishes its lineage to GLaaS.

Every page below is linked as raw markdown. The same content is served as HTML at the equivalent path without the `.md` suffix.

## Get Started

What TReqs is, how it relates to roar and GLaaS, and the setup a new organization needs.

- [Getting started](http://app.treqs.ai/docs.md): Get oriented, and find your way around the docs.
- [roar, GLaaS, and TReqs](http://app.treqs.ai/docs/roar-glaas-treqs.md): Three tools, one lineage story — and which one you reach for when.
- [Organizations](http://app.treqs.ai/docs/organizations.md): Create an organization, or accept an invitation to join one.
- [Projects](http://app.treqs.ai/docs/projects.md): Connect a repository and give training work somewhere to live.
- [Members and roles](http://app.treqs.ai/docs/members-and-roles.md): Adding people to an organization — what each role can do, and how writer and reader seats differ.
- [Compute targets](http://app.treqs.ai/docs/compute-targets.md): Where training runs execute — dedicated machines, on-demand cloud instances, and the secrets they need.

## Training

The core loop: describe a run, get it reviewed, execute it, and publish its lineage.

- [Training requests](http://app.treqs.ai/docs/training-requests.md): The review unit — what goes into one, how it gets approved, and how it moves through its statuses.
- [Workflows](http://app.treqs.ai/docs/workflows.md): The YAML file that defines the stages of a training run.
- [Publishing to GLaaS](http://app.treqs.ai/docs/publishing-to-glaas.md): What a run records, who gets to see it, and how to republish lineage that didn't land.
- [Training runs](http://app.treqs.ai/docs/training-runs.md): Watching queued work execute — statuses, tasks, and logs.

## Extras

Reporting, notifications, and driving TReqs from a terminal.

- [Using the CLI](http://app.treqs.ai/docs/cli.md): Drive TReqs from a terminal — logging in, binding a repo to a project, and running the request lifecycle.
- [Insights](http://app.treqs.ai/docs/insights.md): Training activity and compute utilization, across an organization or a single project.
- [Notifications](http://app.treqs.ai/docs/notifications.md): In-app and Slack notifications — what TReqs tells you about, and how to control it.

## Optional

- [All documentation in one file](http://app.treqs.ai/llms-full.txt): every page above concatenated, for ingesting the whole corpus in one request.