Run agents asynchronously

With asynchronous agents, your team can build and deploy no-code agents in AI Studio and you can use the async applications API to generate content asynchronously or in batches.

For example, your product team can build a no-code agent that creates product description pages, and then you can use the async applications API to generate pages in batches for multiple products.

This guide helps you understand how to run async jobs. The API is similar to the applications API, but it allows you to process batches of requests asynchronously.

​

Endpoint overview

URL: POST https://api.writer.com/v1/applications/{application_id}/jobs

curl 'https://api.writer.com/v1/applications/<application-id>/jobs' \
-X POST \
-H 'Content-Type: application/json' \
-H "Authorization: Bearer $WRITER_API_KEY" \
--data-raw '{
  "inputs": [
    {
      "id": "Input name",
      "value": [
        "Input value"
      ]
    }
  ]
}'

​

Path parameters

​

Request body

The async applications API has the same request body structure as the no-code agents API. It should contain an array of input objects matching the no-code agent’s input schema.

​

Response format

A successful job creation request returns a JSON object with the following structure:

{
  "job_id": "123-456-789",
  "status": "in_progress",
  "created_at": "2024-03-15T10:00:00Z"
}

​

Usage example

The following example demonstrates using the async applications API to generate content asynchronously with a hypothetical product description generation agent.

​

Create and deploy a no-code agent

First, create and deploy a no-code agent in AI Studio. If you don’t already have a no-code agent, follow the text generation guide to get started.

​

Create an async job

Send a POST request with the inputs for the agent to create an async job.

curl 'https://api.writer.com/v1/applications/1234-45090-534590/jobs' \
-X POST \
-H 'Content-Type: application/json' \
-H "Authorization: Bearer $WRITER_API_KEY" \
--data-raw '{
  "inputs": [
    {
      "id": "Product descriptions",
      "value": [
        "Terra running shoe",
        "Aqua swim goggles",
        "Flex yoga mat"
      ]
    }
  ]
}'

​

Check job status

Use the job ID from the creation response to check the status of your job and see the results.

curl 'https://api.writer.com/v1/applications/jobs/<job-id>' \
-H "Authorization: Bearer $WRITER_API_KEY"

The response includes the current status. If the job has completed, the results are in the data.suggestion field, which has the same structure as the no-code agents API.

{
    "id": "123-456-789",
    "status": "completed",
    "application_id": "2932402-23429023894-234234234",
    "created_at": "2025-02-10T18:18:09.501223Z",
    "completed_at": "2025-02-10T18:18:14.470324Z",
    "data": {
        "title": "Social post",
        "suggestion": "# Social post\nImage: A photo of a person running in a pair of Terra running shoes.\n\nCaption:\n\nIntroducing the all-new Terra running shoe, designed to take you further than ever before. With its innovative cushioning system and durable outsole, the Terra is perfect for runners of all levels.\n\nWhether you're hitting the trails or pounding the pavement, the Terra will keep you comfortable and supported mile after mile. So what are you waiting for? Lace up a pair of Terras and start your journey today!\n\n#TerraRunningShoe #RunFurther #NeverStopExploring #GetOutside"
    },
    "error": null
}

​

List all jobs

You can retrieve all jobs for an application to monitor batch processing:

curl 'https://api.writer.com/v1/applications/<application-id>/jobs' \
-H "Authorization: Bearer $WRITER_API_KEY"

​

Retry failed jobs

If a job fails, you can retry it using the retry endpoint:

curl 'https://api.writer.com/v1/applications/jobs/<job-id>/retry' \
-X POST \
-H "Authorization: Bearer $WRITER_API_KEY"

​

Next steps

By following this guide, you can use the async Applications API to handle large-scale tasks.

Next, learn how to enhance your no-code agents with Knowledge Graph, our tool for RAG.