- Create a job to initialize processing
- Upload the input payload separately using a secure pre-signed url
- Poll the job until processing completes and the output is ready, or use webhooks to receive a completion notification
All input and output files are retained for a maximum of thirty (30) days from upload, after which they will be permanently deleted without notice.
How it works
Follow these steps to use job-based APIs:1
Submit the job
Create a job using a Example responseResponse fields
POST request to the workload-specific endpoint. This initializes the asynchronous workflow and returns a unique job_id and a pre-signed S3 url for uploading the input payload.Endpoint pattern2
Upload the input payload
Upload your input payload directly to the pre-signed S3 URL returned in the previous step using a plain HTTP Successful responseOnce the payload is uploaded successfully, the job automatically begins processing.
PUT request.Request requirements- Method:
PUT - Headers:
Content-Type(appropriate for your payload) - Body: Input payload (JSON, CSV, ZIP, etc.)
- No authorization headers required
cURL
3
Get job completion status
After uploading the input payload, the system processes the job asynchronously. You can poll the same endpoint using a Job statusesExample response (succeeded)When the job status is
GET request with the job_id to track progress.If you prefer not to poll, configure webhooks to receive a completion notification instead. See Webhooks. When you receive the webhook, call the same GET /:workload/:function/:sub_function/:job_id endpoint to retrieve the output URL.Endpoint patterncreated— Job created and awaiting uploadqueued— Job queued for processingin_progress— Job is being processedsucceeded— Processing complete and output availablefailed— Job failed during processing
succeeded, use the url field to download the processed result.