> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sharesame.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Job lifecycle

> Understand how ShareSame jobs move from prompt to files.

ShareSame uses a job model instead of a single long request. You start work with `POST /generate`, then poll `GET /job/{job_id}` until the result is ready.

## Status values

### `building`

The job has started, but no usable file output has been returned yet.

### `partial`

Some files are already available. This is useful if you want to preview or stream intermediate results before the job fully finishes.

### `ready`

The file set is complete and ready to deploy.

### `failed`

The job stopped. Read the `error` field and retry with a corrected request.

## Fields returned by `GET /job/{job_id}`

* `status`: current job state
* `files`: array of `{ path, content, language }`
* `error`: plain-language error message when something goes wrong
* `progress`: short progress message
* `elapsed_seconds`: elapsed job time
* `sandbox_session`: sandbox identifier for session reuse
* `deployment_id`: deployment snapshot identifier when available

## Polling pattern

```js theme={null}
const pollUntilReady = async (jobId) => {
  while (true) {
    const response = await fetch(`${process.env.SHARESAME_BASE_URL}/job/${jobId}`, {
      headers: {
        Authorization: `Bearer ${process.env.SHARESAME_API_KEY}`,
      },
    });

    const data = await response.json();

    if (data.status === "ready" || data.status === "failed") {
      return data;
    }

    await new Promise((resolve) => setTimeout(resolve, 2000));
  }
};
```

<Tip>
  Polling every 1 to 2 seconds is usually enough.
</Tip>