Skip to main content
Your agent queues training jobs, and a worker picks them up and runs them. There are four kinds:
  • Browser worker: a tab at train.auremi.ai. It trains only while the tab is open.
  • Desktop app: the same worker as a Windows, Mac, or Linux app. It runs unattended.
  • Python worker: a command-line worker for a GPU machine, with no browser. It runs PyTorch model repos.
  • Cloud compute: SageMaker or Vertex AI in your own cloud account.

​
Pair a browser worker

1

Open the worker

On the training machine, open train.auremi.ai.
2

Request pairing

Enter your Auremi account email and an optional Device name, then click Pair this worker. Copy the eight-character pairing code it shows.
3

Approve it in the console

In the console, open Training Workers from the sidebar. Under Approve a training worker, enter the Pairing code and a Worker name. Leave Eligible jobs on Any compatible job, pick an Authorization length, and click Approve worker.
The worker starts by itself. On Training Workers its status reads available. If the code expires first (the time is shown under it), pair again. Authorization ends after the length you picked. Tick Extend authorization while this worker is healthy to keep it running, or click Renew on the worker later.

​
Install the desktop app

Use the desktop app for a machine that should train with no browser tab open.
Some platforms may not be available yet. Windows comes first.
Open the app and pair it the same way: enter your email, click Pair this worker, and approve the code on Training Workers. Then leave the app running.

​
Run a worker from Python

Use the Python worker on a machine with a GPU that you start from a terminal or a service manager. It trains PyTorch model repos.

Ask your agent, on the training machine

pair prints a pairing code. Approve it on Training Workers, the same as a browser worker. Then keep auremi-worker run running.

​
Train on SageMaker or Vertex

Connect your cloud account once in the console. After that, ask your agent to run jobs there. It builds and uploads the training image itself the first time.

​
SageMaker

1

Open the setup guide

In the console, open Cloud Compute and stay on the GPU tab. Choose SageMaker, click How do I set this up?, and pick Allow Coding Agent To Push Images.
2

Set up AWS

Following the guide:
  • Create an IAM user or role for your coding agent, so it can push the runner image to ECR and submit SageMaker training jobs. Sign the AWS CLI in with it, and check with aws sts get-caller-identity.
  • Create the SageMaker execution role. The guide shows the trust and runtime policy JSON to attach.
3

Fill in the form

  • Display name, AWS account id, and Region
  • Execution role ARN: the execution role you just created.
  • ECR image: <account-id>.dkr.ecr.<region>.amazonaws.com/auremi/runner:latest. This is where your agent pushes the image.
  • The instance type (for example ml.g5.xlarge), Instance count, Max spend, Max runtime minutes, and Storage GB
Click Save SageMaker.
4

Ask your agent

Train on SageMaker

​
Vertex AI

1

Set up Google Cloud

In the console, open Cloud Compute, choose Vertex, and click How do I set this up?. Following the guide, enable the Vertex AI, Artifact Registry, and IAM Service Account Credentials APIs, then create a service account and a Workload Identity Federation provider for Auremi.
2

Fill in the form

  • Display name, GCP project, GCP project number, and Region
  • Service account and Workload identity provider
  • Container image: an Artifact Registry address, such as us-central1-docker.pkg.dev/<gcp-project>/<repo>/auremi-runner:latest. This is where your agent pushes the image.
  • Machine type, Accelerator type, Accelerator count, Max spend, Max runtime minutes, and Boot disk GB
Click Save Vertex.
3

Ask your agent

Train on Vertex

The agent needs the AWS CLI or gcloud signed in on your machine to push the image.

​
Manage workers

Each worker has a card on Training Workers. A worker shows offline when its tab or app is closed or asleep.
  • Drain finishes the current job and takes no new ones.
  • Stop pauses the worker.
  • Revoke removes its access immediately.